834 63 45MB
English Pages 1374 Year 2007
Bundesanstalt für Geowissenschaften und Rohstoffe, Hannover Federal Institute for Geosciences and Natural Resources
Environmental Geology Handbook of Field Methods and Case Studies
This book is based on a joint research project partially funded by the Ministry of Education and Research (BMBF) of the Federal Republic of Germany through the Project Agency Forschungszentrum Karlsruhe GmbH, Water Technology and Waste Management Division (PTKA-WTE / FKZ 0261218). The authors are responsible for the scientific content of their contribution.
Klaus Knödel Gerhard Lange Hans-Jürgen Voigt
Environmental Geology Handbook of Field Methods and Case Studies with contributions by Thekla Abel, Sven Altfelder, Ulrich Beims, Manfred Birke, Norbert Blindow, Antje Bohn, Tilmann Bucher, Reiner Dohrmann, William E. Doll, Dieter Eisenburger, Hagen Hilse, Peter Herms, Bernhard Hörig, Bernhard Illich, Florian Jenn, Stephan Kaufhold, Claus Kohfahl, Franz König, Friedrich Kühn, Manja Liese, Harald Lindner, Reinhard Meyer, Klaus-Henrik Mittenzwey, Kai Müller, Mike Müller, Ranjeet Nagare, Michael Neuhaus, Claus Nitsche, Ricardo A. Olea, Hellfried Petzold, Michael Porzig, Jens Radschinski, Thomas Richter, Knut Seidel, Kathrin R. Schmidt, Dietmar Schmidt, Andreas Schuck, Anke Steinbach, Alejandra Tejedo, Andreas Tiehm, Markus Toloczyki, Peter Weidelt, Thomas Wonik, Ugur Yaramanci
with 501 Figures and 204 Tables
Editor: BUNDESANSTALT FÜR GEOWISSENSCHAFTEN UND ROHSTOFFE Stilleweg 2 30655 HANNOVER GERMANY
ISBN 978-3-540-74669-0 Springer Berlin Heidelberg New York Library of Congress Control Number: 2007937937 This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilm or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer-Verlag. Violations are liable to prosecution under the German Copyright Law. Springer is a part of Springer Science+Business Media springeronline.com © Springer-Verlag Berlin Heidelberg 2007 The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Cover design: deblik, Berlin Production: A. Oelschläger Typesetting: Camera-ready by Claudia Wießner, BGR, Berlin Office Printed on acid-free paper 30/2132/AO 543210
Table of Contents 1
Introduction Klaus Knödel, Gerhard Lange & Hans-Jürgen Voigt
1
2
Preparatory Steps and Common Problems Klaus Knödel, Gerhard Lange & Hans-Jürgen Voigt Placing of Orders and Order Handling Collection and Use of Existing Data Information Campaign and Permit Application Mobilization and Demobilization Land Surveying Quality Assurance and Reporting
11
Remote Sensing Peter Herms, Bernhard Hörig, Friedrich Kühn, Dietmar Schmidt & Anke Steinbach, with a contribution by Tilmann Bucher Aerial Photography Dietmar Schmidt & Friedrich Kühn Principle of the Methods Applications Fundamentals Instruments and Film Survey Practice Interpretation of Aerial Photographs Quality Assurance Personnel, Equipment, Time Needed Examples Photogrammetry Dietmar Schmidt, Peter Herms, Anke Steinbach & Friedrich Kühn Principle of the Methods Applications Fundamentals Instruments Survey Practice Processing and Interpretation of Data Quality Assurance Personnel, Equipment, Time Needed Examples
23
2.1 2.2 2.3 2.4 2.5 2.6 3
3.1 3.1.1 3.1.2 3.1.3 3.1.4 3.1.5 3.1.6 3.1.7 3.1.8 3.1.9 3.2
3.2.1 3.2.2 3.2.3 3.2.4 3.2.5 3.2.6 3.2.7 3.2.8 3.2.9
11 12 14 16 17 21
23 23 24 25 27 34 36 43 46 47 73
73 74 74 81 82 82 85 86 86
VI 3.3
3.3.1 3.3.2 3.3.3 3.3.4 3.3.5 3.3.6 3.3.7 3.3.8 3.3.9 4
4.1 4.1.1 4.1.2 4.1.3 4.1.4 4.1.5 4.1.6 4.1.7 4.1.8 4.1.9 4.2 4.2.1 4.2.2 4.2.3 4.2.4 4.2.5 4.2.6 4.2.7 4.2.8 4.2.9
Table of Contents
Nonphotographic Imaging from Aircraft and Space-borne Platforms Friedrich Kühn, Bernhard Hörig & Dietmar Schmidt, with a contribution by Tilmann Bucher Principle of the Methods Applications Fundamentals Instruments Survey Practice Processing and Interpretation of Data Quality Assurance Personnel, Equipment, Time Needed Examples Geophysics Norbert Blindow, Klaus Knödel, Franz König, Gerhard Lange, Harald Lindner, Reinie Meyer, Klaus-Henrik Mittenzwey, Andreas Schuck, Knut Seidel, Peter Weidelt, Thomas Wonik, Ugur Yaramanci, with contributions by Dieter Eisenburger, Bernhard Illich, Ricardo A. Olea, Hellfried Petzold & Thomas Richter Magnetic Methods Klaus Knödel Principle of the Methods Applications Fundamentals Instruments Survey Practice Processing and Interpretation of the Measured Data Quality Assurance Personnel, Equipment, Time Needed Examples Gravity Methods Knut Seidel & Harald Lindner Principle of the Methods Applications Fundamentals Instruments Survey Practice Processing and Interpretation of the Measured Data Quality Assurance Personnel, Equipment, Time Needed Examples
97
97 99 100 115 124 125 134 136 137 161
161 161 162 162 168 170 171 174 175 176 185 185 186 186 191 192 195 197 198 199
Table of Contents
VII
4.3
205
4.3.1 4.3.2 4.3.3 4.3.4 4.3.5 4.3.6 4.3.7 4.3.8 4.3.9 4.4 4.4.1 4.4.2 4.4.3 4.4.4 4.4.5 4.4.6 4.4.7 4.4.8 4.4.9 4.5
4.5.1 4.5.2 4.5.3 4.5.4 4.5.5 4.5.6 4.5.7 4.5.8 4.5.9 4.5.10 4.6 4.6.1 4.6.2 4.6.3 4.6.3.1 4.6.3.2
Direct Current Resistivity Methods Knut Seidel & Gerhard Lange Principle of the Methods Applications Fundamentals Instruments Survey Practice Processing and Interpretation of Measured Data Quality Assurance Personnel, Equipment, Time Needed Examples Electromagnetic Methods Gerhard Lange & Knut Seidel Principle of the Methods Applications Fundamentals Instruments Survey Practice Processing and Interpretation of Measured Data Quality Assurance Personnel, Equipment, Time Needed Examples Ground Penetrating Radar Norbert Blindow, with contributions by Dieter Eisenburger, Bernhard Illich, Hellfried Petzold & Thomas Richter Principle of the Methods Applications Fundamentals Instruments Survey Practice Processing, Presentation and Interpretation of the Measured Data Quality Assurance Personnel, Equipment, Time Needed Examples Special Applications and New Developments Dieter Eisenburger Seismic Methods Andreas Schuck & Gerhard Lange Principle of the Methods Applications Fundamentals Propagation of Elastic Waves Elastic Parameters and Seismic Velocities
205 207 207 215 216 221 225 227 228 239 239 243 243 255 260 263 266 269 270 283
283 285 286 297 300 302 303 304 305 316 337 337 340 341 341 343
VIII 4.6.3.3 4.6.3.4 4.6.3.5 4.6.4 4.6.4.1 4.6.4.2 4.6.4.3 4.6.5 4.6.5.1 4.6.5.2 4.6.5.3 4.6.5.4 4.6.6 4.6.6.1 4.6.6.2 4.6.6.3 4.6.6.4 4.6.6.5 4.6.7 4.6.7.1 4.6.7.2 4.6.7.3 4.6.7.4 4.6.7.5 4.6.7.6 4.6.7.7 4.6.7.8 4.6.8 4.7 4.7.1 4.7.2 4.7.3 4.7.4 4.7.5 4.7.6 4.7.7 4.7.8 4.7.9 4.8 4.8.1 4.8.2 4.8.3
Table of Contents
Reflection, Transmission and Diffraction Surface Waves Seismic Resolution Instruments Seismic Sources Seismic Sensors Seismic Recording Instruments Seismic Refraction Surveying Principle of the Method Survey Practice Processing and Interpretation Personnel, Equipment, Time Needed Seismic Reflection Surveying Principle of the Method Survey Practice Processing and Interpretation of Measured Data Quality Assurance Personnel, Equipment and Time Needed Borehole Seismic Methods Principle of the Methods Applications Fundamentals Instruments Survey Practice Processing and Interpretation of Measured Data Quality Assurance Personnel, Equipment, Time Needed Examples Surface Nuclear Magnetic Resonance Gerhard Lange, Ugur Yaramanci & Reinhard Meyer Principle of the Method Applications Fundamentals Instruments Survey Practice Processing and Interpretation of the Measured Data Quality Assurance Personnel, Equipment, Time Needed Examples Borehole Logging Thomas Wonik with a contribution by Ricardo A. Olea Principle of the Methods Applications Slimhole Logging Equipment and Logging Methods
347 350 352 354 354 359 362 363 363 364 365 369 369 369 373 376 382 383 384 384 384 385 386 386 387 387 388 388 403 403 404 404 408 410 413 415 416 416 431 431 433 434
Table of Contents
4.8.3.1 4.8.3.2 4.8.3.3 4.8.3.4 4.8.3.5 4.8.3.6 4.8.3.7 4.8.4 4.8.5 4.8.6 4.8.7 4.9
4.9.1 4.9.2 4.9.3 4.9.3.1 4.9.3.2 4.9.3.3 4.9.4 4.9.5 4.9.6 4.9.7 4.9.8 4.9.9 5
5.1
5.1.1 5.1.2
IX
Radioactivity Logging Methods Electrical Methods Electromagnetic Methods Acoustic Methods Optical Methods Methods for Determining the Properties of Drilling Fluids (Fluid Logs) Methods for Determining Borehole Properties Survey Practice, Personnel, Equipment, Time Needed Quality Assurance Processing and Interpretation of the Logging Data and Examples Expected Future Developments Geophysical In-situ Groundwater and Soil Monitoring Franz König, Klaus Knödel, Klaus-Henrik Mittenzwey & Peter Weidelt Principle of the Methods Applications Fundamentals Environmental Parameters Optical Spectroscopy EM Monitoring Instruments Field work Processing Quality Assurance Personnel, Equipment, Time Needed Examples
436 438 440 440 441 441
Geological, Hydrogeological, Geochemical and Microbiological Investigations Sven Altfelder, Ulrich Beims, Manfred Birke, Reiner Dohrmann, Hagen Hilse, Florian Jenn, Stephan Kaufhold, Klaus Knödel, Claus Kofahl, Manja Liese, Kai Müller, Mike Müller, Ranjeet Nagare, Michael Neuhaus, Claus Nitsche, Michael Porzig, Jens Radschinski, Katrin R. Schmidt, Andreas Thiem & Hans-Jürgen Voigt Methods for Characterizing the Geological Setting Klaus Knödel, Kai Müller, Michael Neuhaus, Florian Jenn & Hans-Jürgen Voigt Geologic Field Observations Florian Jenn, Klaus Knödel & Hans-Jürgen Voigt Trenching Klaus Knödel & Hans-Jürgen Voigt
507
443 443 447 449 470 475
475 477 478 478 481 484 488 492 493 494 494 495
507
511 519
X 5.1.3 5.1.4 5.2
5.2.1
5.2.2
5.2.3
5.2.4
5.2.5
5.2.6 5.2.7
5.2.7.1
5.2.7.2 5.2.7.3
5.3
Table of Contents
Drilling Florian Jenn & Hans-Jürgen Voigt Direct Push Technology Kai Müller, Michael Neuhaus & Hans-Jürgen Voigt Methods for Characterizing the Hydrologic and Hydraulic Conditions Ulrich Beims, Florian Jenn, Klaus Knödel, Manja Liese, Ranjeet Nagare, Claus Nitsche, Michael Porzig & Hans-Jürgen Voigt Precipitation Florian Jenn, Klaus Knödel, Manja Liese & Hans-Jürgen Voigt Evaporation and Evapotranspiration Florian Jenn, Klaus Knödel, Manja Liese & Hans-Jürgen Voigt Runoff Florian Jenn, Klaus Knödel, Manja Liese & Hans-Jürgen Voigt Infiltration Florian Jenn, Klaus Knödel, Manja Liese & Hans-Jürgen Voigt Groundwater Recharge Florian Jenn, Klaus Knödel, Manja Liese & Hans-Jürgen Voigt Groundwater Monitoring Florian Jenn, Claus Nitsche & Hans-Jürgen Voigt Determination of Hydraulic Parameters Ulrich Beims, Ranjeet Nagare, Claus Nitsche, Michael Porzig & Hans-Jürgen Voigt Infiltrometer and Permeameter Tests Florian Jenn, Ranjeet Nagare, Michael Porzig & Hans-Jürgen Voigt Pumping Tests Ulrich Beims Laboratory Methods Florian Jenn, Claus Nitsche & Hans-Jürgen Voigt
524
Methods for Characterizing the Geochemical and Microbiological Conditions Sven Altfelder, Manfred Birke, Reiner Dohrmann, Hagen Hilse, Florian Jenn, Stephan Kaufhold, Klaus Knödel, Claus Nitsche, Kathrin R. Schmidt, Andreas Thiem & Hans-Jürgen Voigt
749
540 567
569
581
590
603
612
619 643
649
681 711
Table of Contents
Sampling and Analysis of Groundwater and Surface Water Claus Nitsche & Hans-Jürgen Voigt 5.3.1.1 Planning and Preparation of Work 5.3.1.2 Groundwater Sampling 5.3.1.3 Groundwater Analysis Klaus Knödel 5.3.2 Sampling and Analysis of Soil, Rock, Stream and Lacustrine Sediments Manfred Birke, Claus Nitsche & Hans-Jürgen Voigt 5.3.2.1 Planning and Preparation of Work 5.3.2.2 Sampling of Soil, Rock, Stream and Lacustrine Sediments 5.3.2.3 Analysis of Soil, Rock, Stream and Lacustrine Sediments 5.3.3 Sampling and Analysis of Soil Gas and Landfill Gas Hagen Hilse & Hans-Jürgen Voigt 5.3.3.1 Planning and Preparation of Work 5.3.3.2 Sampling and Analysis of Soil Gas and Landfill Gas 5.3.4 Methods for Chemical Analysis used in Geochemical Investigations Florian Jenn 5.3.5 Laboratory Methods for the Determination of Migration Parameters Sven Altfelder with a contribution by Claus Nitsche 5.3.5.1 Basic Theory of Sorption 5.3.5.2 Basic Theory of Transport 5.3.5.3 Sampling and Preparation of Soil or Sediment for the Determination of Migration Parameter Values 5.3.5.4 Batch Tests 5.3.5.5 Column Experiments 5.3.5.6 Clays and Clay Minerals Reiner Dohrmann 5.3.5.7 Cation Exchange Capacity Reiner Dohrmann 5.3.5.8 Carbonates Stephan Kaufhold 5.3.5.9 Iron and Manganese Oxides Stephan Kaufhold 5.3.5.10 Organic Carbon Klaus Knödel 5.3.6 Methods to Evaluate Biodegradation at Contaminated Sites Andreas Thiem & Kathrin R. Schmidt 5.3.6.1 Microbial Processes in the Subsurface 5.3.6.2 Assessment Methods 5.3.6.3 Case Studies 5.3.1
XI 758 759 768 778 785
786 790 797 807 810 811 816
823
824 830 831 832 838 849 853 862 867 871 876 877 883 892
XII 5.4
5.4.1 5.4.1.1 5.4.1.2 5.4.1.3 5.4.1.4 5.4.1.5 5.4.2 5.4.3 5.4.3.1 5.4.3.2 5.4.3.3 5.4.4 5.4.4.1 5.4.4.2 6
6.1
6.1.1 6.1.2 6.1.2.1 6.1.2.2 6.1.2.3 6.1.2.4 6.1.3 6.2
Table of Contents
Interpretation of Geological, Hydrogeological, and Geochemical Results Florian Jenn, Claus Kofahl, Mike Müller, Jens Radschinski & Hans-Jürgen Voigt Statistical Methods Florian Jenn & Hans-Jürgen Voigt Univariate Statistics Multivariate Statistics Time Series Analysis Geostatistics and Interpolation of Spatial Data Specific Tests of Hydrogeochemical Data Conceptual Model Hans-Jürgen Voigt & Jens Radschinski Groundwater Flow Modeling Claus Kofahl Fundamentals of Groundwater Flow Modeling Programs Guide for Construction and Use of a Groundwater Model Contaminant Transport Modeling Mike Müller Fundamentals of Transport Modeling Model Application Integration of Investigation Results Thekla Abel, Manfred Birke, Antje Bohn, Klaus Knödel, Gerhard Lange, Alejandra Tejedo, Markus Toloczyki & Ugur Yaramanci Data Fusion Klaus Knödel, Marcus Toloczyki, Antje Bohn, Thekla Abel, Gerhard Lange & Alejandro Tejedo Reprocessing and New Data Presentation Klaus Knödel & Gerhard Lange Geographic Information Systems Marcus Toloczyki, Thekla Abel & Alejandra Tejedo Fundamentals Hardware, Network, Software, and Manpower Data Acquisition and Analysis Examples Other Data Fusion Examples Klaus Knödel & Antje Bohn Joint Interpretation Manfred Birke, Klaus Knödel, Gerhard Lange & Ugur Yaramanci
941
941 942 952 958 959 960 962 1001 1002 1009 1014 1020 1021 1041 1053
1054
1055 1061 1063 1067 1072 1075 1091 1099
Table of Contents
6.2.1 6.2.2
6.2.2.1
6.2.2.2 6.2.2.3
Qualitative and Semiquantitative Approach Klaus Knödel & Gerhard Lange Quantitative Approach Manfred Birke, Klaus Knödel, Gerhard Lange & Ugur Yaramanci Joint Quantitative Interpretation of Several Geophysical Measurements and Core Analysis Results Ugur Yaramanci & Gerhard Lange Joint Inversion Klaus Knödel, Ugur Yaramanci & Gerhard Lange Joint Interpretation Using Statistical Methods Manfred Birke Glossary Abbreviations Units of Measure SI Prefixes None SI Units Physical Constants Mathematical Symbols and Constants Subject Index
XIII 1099 1150
1050
1059 1063
1195 1319 1333 1335 1335 1336 1337 1339
List of Authors Dr. Thekla Abel Bundesanstalt für Geowissenschaften und Rohstoffe Stilleweg 2 D-30655 Hannover Dr. Sven Altfelder Bundesanstalt für Geowissenschaften und Rohstoffe Stilleweg 2 D-30655 Hannover Prof. Dr.-Ing. Ulrich Beims GFI Grundwasserforschungsinstitut GmbH Dresden Meraner Straße 10 D-01217 Dresden Dr. Manfred Birke Bundesanstalt für Geowissenschaften und Rohstoffe Dienstbereich Berlin Wilhelmstraße 25-30 D-13593 Berlin Dr. Norbert Blindow Westfälische Wilhelms-Universität Institut für Geophysik Corrensstraße 24 D-48149 Münster Antje Bohn Brandenburgische Technische Universität Cottbus LS Umweltgeologie Erich-Weinert-Straße 1 D-03406 Cottbus Tilmann Bucher Deutsches Zentrum für Luft- und Raumfahrt Berlin-Adlershof Einrichtung Optische Informationssysteme am Institut für Robotik und Mechatronik Rutherfordstraße 2 D-12489 Berlin
List of Authors
Dr. Reiner Dohrmann Bundesanstalt für Geowissenschaften und Rohstoffe Stilleweg 2 D-30655 Hannover Dr. William Doll Research Leader Batelle - Oak Ridge Operations 105 Mitchell Road, Ste 103 Oak Ridge, TN 37830 Dieter Eisenburger Bundesanstalt für Geowissenschaften und Rohstoffe Stilleweg 2 D-30655 Hannover Dr.-Ing. Hagen Hilse GICON – Großmann Ingenieur Consult GmbH Tiergartenstraße 48 D-01219 Dresden Peter Herms Hansa Luftbild AG Elbestraße 5 D-48145 Münster Bernhard Hörig Bundesanstalt für Geowissenschaften und Rohstoffe Dienstbereich Berlin Wilhelmstraße 25-30 D-13593 Berlin Bernhard Illich GGU Gesellschaft für Geophysikalische Untersuchungen Amalienstraße 4 D-76133 Karlsruhe Florian Jenn Brandenburgische Technische Universität Cottbus LS Umweltgeologie Erich-Weinert-Straße 1 D-03406 Cottbus
XV
XVI Dr. Stephan Kaufhold Bundesanstalt für Geowissenschaften und Rohstoffe Stilleweg 2 D-30655 Hannover Dr. Klaus Knödel Bundesanstalt für Geowissenschaften und Rohstoffe Dienstbereich Berlin Wilhelmstraße 25-30 D-13593 Berlin Dr. Claus Kohfahl FU Berlin Institut für Geologische Wissenschaften Fachrichtung Geochemie, Hydrologie, Mineralogie Arbeitsbereich Hydrologie Malteserstrasse 74-100, Haus B D-12249 Berlin Franz König Bundesanstalt für Geowissenschaften und Rohstoffe Dienstbereich Berlin Wilhelmstraße 25-30 D-13593 Berlin Dr. Friedrich Kühn Bundesanstalt für Geowissenschaften und Rohstoffe Stilleweg 2 D-30655 Hannover Gerhard Lange Bundesanstalt für Geowissenschaften und Rohstoffe Dienstbereich Berlin Wilhelmstraße 25-30 D-13593 Berlin Manja Liese Brandenburgische Technische Universität Cottbus LS Umweltgeologie Erich-Weinert-Straße 1 D-03406 Cottbus
List of Authors
List of Authors
Prof. Dr. Harald Lindner TU Bergakademie Freiberg Institut für Geophysik Gustav-Zeuner-Straße 12 D-09599 Freiberg Reinhard Meyer Groundwater Sciences Research Group Natural Resources and the Environment (NRE), Council for Scientific and Industrial Research (CSIR) PO Box 395 Pretoria, 0001 Dr. Klaus-Henrik Mittenzwey OPTOSENS Optische Spektroskopie und Sensortechnik GmbH Rudower Chaussee 29 (IGZ) D-12489 Berlin Kai Müller Boden- und Grundwasserlabor GmbH (BGD) Tiergartenstrasse 48 D-01219 Dresden Dr. Mike Müller Ingenieurbüro für Grundwasser GmbH Nonnenstrasse 9 D-04229 Leipzig Ranjeet Nagare, M.Sc. Brandenburgische Technische Universität Cottbus LS Umweltgeologie Erich-Weinert-Straße 1 D-03406 Cottbus Dr. Michael Neuhaus FUGRO Consult GmbH Wolfener Straße 36 V D-12681 Berlin Dr. Claus Nitsche Boden- und Grundwasserlabor GmbH (BGD) Tiergartenstrasse 48 D-01219 Dresden
XVII
XVIII Ricardo A. Olea Bundesanstalt für Geowissenschaften und Rohstoffe, Dienstbereich Berlin Wilhelmstraße 25-30 D-13593 Berlin
List of Authors
Now: Eastern Energy Resorces US Geological Survey 12201 Sunrise Valley Dr., MS956 Reston, VA20192
Dr. Hellfried Petzold LAUBAG Lausitzer Braunkohle AG Knappenstraße 1 D-01968 Senftenberg Michael Porzig Brandenburgische Technische Universität Cottbus LS Umweltgeologie Erich-Weinert-Straße 1 D-03406 Cottbus Jens Radschinski Brandenburgische Technische Universität Cottbus LS Umweltgeologie Erich-Weinert-Straße 1 D-03406 Cottbus Dr. Thomas Richter Bo-Ra-Tec GmbH Damaschkestr. 19a D-99425 Weimar Knut Seidel Geophysik GGD Gesellschaft für Geowissenschaftliche Dienste Ehrensteinstraße 33 D-04105 Leipzig
Now: GGL Geophysik und Geotechnik Leipzig GmbH Bautzner Straße 67 D-04347 Leipzig
Kathrin R. Schmidt DVGW - Technologiezentrum Wasser (TZW) Karlsruher Straße 84 76139 Karlsruhe
XIX
List of Authors
Dietmar Schmidt Hansa Luftbild AG Elbestraße 5 D-48145 Münster Dr. Andreas Schuck Geophysik GGD Gesellschaft für Geowissenschaftliche Dienste Ehrensteinstraße 33 D-04105 Leipzig
Now: GGL Geophysik und Geotechnik Leipzig GmbH Bautzner Straße 67 D-04347 Leipzig
Anke Steinbach Hansa Luftbild AG Elbestraße 5 D-48145 Münster Alejandra Tejedo SEGEMAR Servicio Geológico Minero Argentino Buenos Aires Argentina Dr. Andreas Tiehm DVGW - Technologiezentrum Wasser (TZW) Karlsruher Straße 84 D-76139 Karlsruhe Dr. Markus Toloczyki Bundesanstalt für Geowissenschaften und Rohstoffe Stilleweg 2 D-30655 Hannover Prof. Dr. Hans-Jürgen Voigt Brandenburgische Technische Universität Cottbus LS Umweltgeologie Erich-Weinert-Straße 1 D-03406 Cottbus Prof. Dr. Peter Weidelt TU Braunschweig Institut für Geophysik und Extraterrestrische Physik Mendelsohnstraße 3 D-38106 Braunschweig
XX Dr. Thomas Wonik GGA Institut für Geowissenschaftliche Gemeinschaftsaufgaben Stilleweg 2 D-30655 Hannover Prof. Dr. Ugur Yaramanci TU Berlin, Fakultät 6 Fachbereich Bauingenieurwesen und Angewandte Geophysik Ackerstraße 71-76 D-13355 Berlin
List of Authors
List of Reviewers Dr. Ron D. Barker Earth Sciences The University of Birmingham Edgbaston Birmingham B15 2TT United Kingdom Jean Bernard IRIS instruments 1 Avenue Buffon, BP 6007 45060 Orleans cedex 02 France Dr. Manfred Birke Federal Institute for Geosciences and Natural Resources, DB Berlin Wilhelmstraße 25-30 D-13593 Berlin Prof. Dr. Hans Jürgen Burkhardt Technical University Berlin Dept. of Applied Geophysics Ackerstrasse 71-76 D-13355 Berlin Dr. Christoph Grissemann Federal Institute for Geosciences and Natural Resources, DB Berlin Stilleweg 2 D-30655 Hannover Dr. Martin Herold ESA GOFC GOLD PROJECT OFFICE Department of Geography Friedrich Schiller University Jena Loebdergraben 32 D-07743 Jena
XXII
List of Reviewers
Prof. Dr. Anatoly Legchenko Institut de Recherche pour la Developpement LTHE, BP53 38041 Grenoble cedex 9 France Prof. Dr. Ludwig Luckner Dresdner Grundwasserforschungszentrum e.V. Meraner Str. 10 D-01217 Dresden Reinhard Meyer Groundwater Sciences Research Group Natural Resources and the Environment CSIR Pretoria, 0001 South Africa Dr. Edgar Stettler Council for Geoscience Pretoria, 0001 South Africa Dr. Manfred Teschner Federal Institute for Geosciences and Natural Resources Stilleweg 2 D-30655 Hannover Prof. Dr. Peter Weidelt Technical University Braunschweig Institut of Geophysics und Meteorology Mendelssohnstraße 3 D-38106 Braunschweig
Now: Thani Mineral Investment, United Arab Emirates
Acknowledgements This handbook is a product of the Thai-German Research Project WADIS (Waste Disposal: Recommendations for Site Investigations of Waste Disposal Sites and Contaminated Sites in Thailand, funded by the Ministry of Education and Research (BMBF) of the Federal Republic of Germany (grant no. 0261218) through its Project Management Agency Forschungszentrum Karlsruhe, Water Technology and Waste Management Division (PTKAWTE). The editors and authors gratefully acknowledge the financial support by the German Federal Ministry of Education and Research (BMBF) and the scientific and administrative support by the Project Management Agency. Without the assistance of Prof. Dr. D. Mager, Federal Ministry of Economics and Technology, Mineral Resources and Geosciences Section, the project would not have been started and successfully carried out. The WADIS project was approved by the Thai Cabinet on 30th May 2000 and the Project Agreement was signed in September 2000 in Bangkok between the Ministry of Industry and the German Federal Institute for Geosciences and Natural Resources (BGR). Strongly appreciated are, in particular, the support and efforts continually given to the research project by the management of the Department of Mineral Resources (DMR) and especially the director of the Environmental Geology and Geohazards Division, Khun Worawoot Tantiwanit, and his colleagues in Bangkok and in the Chiang Mai office. An important aspect of the project work was the excellent team spirit among Thai and German partners. The joint Thai-German project group included the following institutions and companies: DMR - Department of Mineral Resources, Bangkok
BGR - Federal Institute for Geosciences and Natural Resources, Hannover
CMU - Chiang Mai University, Chiang Mai
BTU - Brandenburg Technische Universität, Cottbus
ATOP Technology Co. Ltd., Bangkok
GGD - Geophysik GmbH, Leipzig
PCD - Pollution Control Department, Bangkok
BGD - Soil and Groundwater Laboratory GmbH, Dresden
Sky Eyes Co. Ltd., Bangkok
Hansa Luftbild GmbH, Münster
AZTEC Engineering Co. Ltd., Lampang
Gicon GmbH, Dresden
METRIX Associates Co. Ltd., Bangkok
Ingenieurbüro Sehlhoff GmbH, Vilsbiburg
XXIV
Acknowledgements
The contributions and achievements of all partners in these institutions and companies are gratefully acknowledged. This book is a joint venture of scientists in companies, universities, and institutions. Thanks are also due to the colleagues, companies and institutions outside the WADIS project group whose contributions and case studies enriched this handbook. Dr. Christoph Grissemann of the Geophysics for Resource Management Section of the Federal Institute for Geosciences and Natural Resources conducted the final phase of the WADIS project. The editors would like to thank him for the excellent cooperation and beneficial discussions during the preparation of this handbook. Particular thanks go to all reviewers (see List of Reviewers) for the thorough revision of manuscripts and numerous constructive comments for improvement. Dr. R. Clark Newcomb has linguistically revised most of the contributions to this book. The manuscripts have benefited from his helpful comments and reviews. Henry Toms was an indispensable help in geoscientific terminology. The authors wish to thank both colleagues for their support. Last but not least, the authors and editors are indebted to Claudia Wießner and Ingrid Boller for their careful preparation of figures and layout of the handbook. They contributed substantially to the design and quality of this book. Thanks also to Claudia Kirsch for her assistance in the handbook preparation.
Preface As earth’s population continues to grow and the detrimental aftereffects of industrialization and environmental negligence become more apparent, society has become more aware of, and concerned about, stewardship of the natural environment – water, soil, and air. Sustainable development has become more widely received and promoted in many parts of the world. The need is now critical for earth and environmental scientists and engineers to work together to implement technologies that can preserve our environment. The Earth’s population was 6.6 billion as of April 2007 according to the U.S. Census Bureau. This number is expected to rise to 9.4 billion by 2050. The population is increasing the demand for natural resources and energy, and increasing stress on the environment. Thus, protection of the environment and remediation of damage to the environment must be a priority. It is also important to develop procedures that will help to avert further damage to the environment and to recognize as early as possible the risks associated with changes in the environment. Many methodologies and technologies have become more advanced in the past few decades, and new technologies and approaches have been developed, all to address the growing need for environmental assessment, monitoring, and remediation. As these technologies have grown, the need for interdisciplinary cooperation has also become more apparent. Specialists in remote sensing, geophysical methods, hydrogeology, geology, and geochemistry must maintain current awareness of developments within their sister disciplines in order to formulate effective overall approaches for environmental issues. Too often, resolution of environmental problems is constrained by political or economic boundaries. In many parts of the world, standards for acceptable air, water, and soil quality must be compromised in order to meet more pressing human needs. It is therefore important that cost-effective interdisciplinary solutions be developed, in order to allow all nations and socio-economic groups to benefit from a clean, sustainable environment. Economic development is ultimately limited by environmental quality. Water is needed for all aspects of life. All social and economic activities depend on having a reliable supply of high quality water. As populations grow and economic activity increases, many countries are rapidly confronted with the problem of water scarcity, which limits economic development. More than 1 billion people, mainly in the developing countries, lack an adequate supply of safe drinking water. By 2050, 25 percent of the people on Earth will live in countries in which water is permanently scarce. Contaminated drinking water is a major cause of disease and death in developing countries. An adequate water supply is a prerequisite for human existence, not only for drinking, but also for agriculture. Degradation of soils also poses a direct threat to food production in developing countries. The availability of arable land is decreasing. Population growth is complicating the situation. According to the
XXVI
Acknowledgements
United Nations Environment Programme (UNEP), there are more refugees from a deteriorating environment than from war. Others warn that water quality will play a growing role in regional conflicts and wars. High quality water and uncontaminated soil must be given greater priority throughout the world. This book, Environmental Geology - Handbook of Field Methods and Case Studies, is intended to enable progress toward these challenging goals. It provides a broad spectrum of investigatory methods in several disciplines to support cross-fertilization among experts in those disciplines and others. Methods that are treated in this book include remote sensing, geophysics, geology, hydrogeology, geochemistry, and microbiology. Most of the methods described in this handbook are available and used in developing countries. Information is provided about the principle of the method, possible applications, fundamentals, instruments, survey practice, processing and interpretation of the data, quality assurance, personnel, equipment, and time needed. Examples are given, as well as references and sources for further reading. Besides geoscientific methods, the procedures for stepwise site investigations are described, as well as common problems encountered in field operations. This handbook is not intended to be used as a textbook, but instead as a reference, providing insights into the fundamentals, application and limits of methods. Interdisciplinary case studies from different parts of the world have been selected as examples for extrapolation to other geoenvironmental concerns. With this structure the handbook can be used as a practical guide for training students. The descriptions of the methods and case studies also illustrate the advantages of interdisciplinary geoscientific site investigations to decisionmakers who deal with environmental investigations. This applies to both remediation assessments and preventative measures. Thorough and knowledgeable application of such an approach will enhance its reliability, credibility, and value to future generations. William E. Doll Oak Ridge, 2007
Environmental Geology, 1 Introduction
1
1 Introduction KLAUS KNÖDEL, GERHARD LANGE & HANS-JÜRGEN VOIGT
Increasing population density and industrialization are creating a high strain on the natural environment and resources of many countries. Therefore, precautionary measures to protect the environment and remedial action to repair the damages of the past have high priority. Resources to be protected are surface water and groundwater, soil and air. Hazards to these resources are landfills and industrial sites as well as mining facilities, including tailings, conditioning plants, and smelters, oil refineries, distribution facilities and pipelines, gas stations and other areas used by humans (e.g., military training sites). Waste disposal, mining, and industrial sites are an absolutely necessary part of the infrastructure of an industrial society. Suitable new sites must be found for the disposal of waste and for mining and industrial facilities. It is often very difficult to obtain political approval and this is possible only if state-of-the-art methods are used to show that such sites have layers that can function as barrier, preventing entry of contaminants into the environment. Areas of both consolidated and unconsolidated rock can be suitable sites for landfills and industrial facilities. Knowledge and experience with the disposal of waste and the operation of mines and industrial facilities in an ecologically nondetrimental way have been acquired only gradually during the past several decades. On the basis of this knowledge, numerous abandoned landfills, mining, and industrial sites must now be regarded as hazardous. Impermeable layers at such sites are the most important barrier for impeding the spread of pollutants. It must be assumed that this geological barrier is always of importance, since the currently used techniques for preparing sites for landfills, mines and industrial plants will prevent the spread of pollutants for only a finite time. A site investigation stepwise is usually carried out in at least two phases: (1) a orientating investigation and (2) a detailed investigation. A flow chart of the stepwise procedure for site investigations is shown in Figure 1-1. Some references to chapters of this book are given in the flow chart. Field and laboratory work as well as data processing, data presentation and interpretation of the data from individual methods are described in the
2
1 Introduction
chapters of Parts 3 to 5. Solutions to common problems are treated in Part 2. For more details of site investigation procedure see books with strategies and recommendations for site investigations (e.g., WILKEN & KNÖDEL, 1999, VOIGT et al., 2006).
Fig. 1-1: Flow chart for site investigations
Environmental Geology, 1 Introduction
3
In the orientating investigation, the following information is obtained from maps and other archived data sources: x topography, land use and vegetation, settlements, roads and railways, x climate: precipitation, temperature, evapotranspiration, direction and velocity of the wind, as well as the frequency of strong winds, x hydrological and hydrogeological conditions: streams, lakes and ponds, springs, wells, use and quality of surface and groundwater, runoff, water balance, aquifer/aquiclude properties and stratigraphy, groundwater table, groundwater recharge and discharge, x geology: soil, geological structures, stratigraphy and lithology, x ecological aspects: e.g., nature reserves, protected geotopes, water protection areas. This is accompanied by a reconnaissance survey in the field and by a historical review of earlier use of the site (interviews of persons who lived or worked around the site during the time of mining or industrial operations). The following aspects or parts of them must be taken into account for a detailed site investigation and assessment: x geology: thickness and lateral extent of strata and geological units, lithology, homogeneity and heterogeneity, bedding conditions and tectonic structures, fractures, impact of weathering, x groundwater: water table, water content, direction and rate of groundwater flow, hydraulic conductivity, value of aquifer, x geochemical site characterization: chemical composition of soil, rocks and groundwater, estimation of contaminant retention, x geotechnical stability: The geological barrier must be capable of adsorbing strain from the weight of a landfill, slag heap or industrial building. x geogenic events: active faults, karst, earthquakes, subsidence, landslides, x anthropogenic activities: mining damage, buildings, quarries, gravel pits, clay pits, etc., and x changes in soil and groundwater quality. An interdisciplinary geoscientific program is required for a site investigation. Numerous methods are available for such studies. The geological and hydrogeological conditions as well as the surface conditions (e.g., vegetation, surface sealing, buildings) at the site must be taken into consideration when the investigation methods are selected. There are some rules of thumb for site investigations: Start with the less expensive methods, expanding as necessary to more expensive methods for detailed investigations, i.e., remote sensing
4
1 Introduction
before geophysics, geology and hydrogeology before geochemistry and modeling. Mapping should be carried out before sounding and drilling, investigations of the area as a whole before point data. First, a representation of the data is made, it is then interpreted, and an assessment of the conditions at the site is made. Not only the geological structures immediately below a landfill, slag heap or industrial site has to be examined, but also the surrounding area. The investigation of the surrounding area must include that part of the geological barrier that is expected to be needed for contaminant retention and that part of the regional groundwater system that will possibly become contaminated. Each landfill, mining or industrial site, whether in operation or abandoned, is within a groundwater system of several tens of square kilometers. A general survey of this area has to be made. A detailed study must be carried out in an area of 0.1 - 1 km² around the site itself. The structures down to a depth of 50 m are relevant to a study of abandoned and planned waste disposal sites. It is often necessary to extend the investigations down to 150 m to obtain information on the groundwater system. In some cases, the geology and the groundwater system below groundwater protection areas also have to be investigated in order to assess their vulnerability to pollution. The following methods and tools can be used to assess a geological barrier and the potential for the spread of contaminants: Remote sensing methods can provide geoscientific data for large areas in a relatively very short time. They are not limited by extremes in terrain or hazardous conditions that may be encountered during an on-site appraisal. In many cases aerial photographs and satellite images should be used to prepare a base map of the investigation area. Remote sensing methods can enable a preliminary assessment and site characterization of an area prior to the use of more costly and time consuming techniques, such as field mapping, geophysical surveys and drilling. The data obtained from satellite-based remote-sensing systems is best suited for regional studies as well as for detecting and monitoring large-scale environmental problems. However, for detailed site characterization, satellite data is sometimes of limited use due to relatively low spatial resolution. Mapping scales of 1 : 10 000 or larger are required for a detailed geoenvironmental assessment of landfills, mining, and industrial sites. Highresolution aerial photographs, airborne scanners, and some satellite-based remote-sensing systems provide data at the required spatial resolution (e.g., 70 cm and better). Aerial photographs made at different times can reveal the changes at sites suspected to be hazardous. Geophysical methods are used to develop a model of the geology below the site, to locate fracture zones, to investigate the groundwater system, to detect and delineate abandoned landfills and contamination plumes, as well as to obtain information on the lithology and physical parameters of the ground. Necessary condition for a meaningful use of geophysical methods is the existence of contrasts in the physical parameter values (magnetization,
Environmental Geology, 1 Introduction
5
susceptibility, density, electrical resistivity, seismic velocities, etc.) of soil and rock. The parameter values to be expected at the site must be considered before a geophysical survey is conducted. Geophysical methods supplement each other because they are sensitive for different physical parameters. Seismic methods are used to investigate structures and lithology. Electrical and electromagnetic methods are very sensitive to changes in electrolyte concentrations in the pore water. Ground-penetrating radar can be used in areas with dry, low-conductivity rocks. Both magnetic and electromagnetic mapping have proved useful for locating and delineating concealed landfills. Both methods are fast and easy to conduct, enabling large areas to be investigated in a short time. Seismic, dc-resistivity, electromagnetic and gravity methods are used to investigate groundwater systems on a regional scale. Geophysical surveys help find suitable locations for drilling groundwater observation wells and provide information between boreholes and for areas where it is impossible to drill. Well logging is absolutely necessary. Logging data are not only necessary for processing and interpretation of surface geophysical data but also as a bridge between geophysical surveys and hydrogeological modeling. Geological and hydrogeological studies are used to investigate lithological structures, to determine the homogeneity of the rock, to locate fractures, to determine the permeability of the rock with respect to water, gases and various contaminants, to assess the mechanical stability of the ground, and to obtain data on the groundwater system. Flow and transport models must be developed to estimate groundwater recharge and the potential for groundwater contamination. The main tasks of geological and hydrogeological surveys are to gain information directly by examining outcrops, digging trenches and drilling boreholes, conducting hydraulic tests (e.g., pumping tests and tracer tests) in wells to determine hydraulic properties in situ. This work is augmented by geological mapping, examination of drill cores, construction and expansion of a network of groundwater observation wells. Rock, soil and groundwater samples are taken to determine physical, chemical, petrographic and mineralogical parameters. Special laboratory experiments can be carried out to estimate migration parameters and the texture of rock and soil samples. Data from cone penetration tests and other field and laboratory methods are used to assess the stability of the ground. Geophysical and geochemical methods can be used to delineate and monitor operating and abandoned landfills as well as to determine the spread of contaminants. Geochemical methods are necessary to obtain information about the capacity of the rock to retain contaminants seeping from a landfill, as well as about the degradation of the hazardous substances. The objectives of the site investigation must be well defined. Additionally important is for suitable methods to be chosen to accomplish the objectives. The sections “Possible Applications” and “Examples” in the chapters of Parts 3 to 5 of this book can aid the selection of suitable methods. Suitable
6
1 Introduction
methods are recommended in Tables 1-1 to 1-3, for example, for geological investigations, investigations of landfill waste bodies, and the evaluation of groundwater conditions. These tables should not be considered as a substitute for the advice of an experienced specialist. This handbook is designed to provide geoscientific methods to investigate landfills and mining and industrial sites. It describes methods in the fields of remote sensing, geophysics, geology, hydrogeology, geochemistry, and microbiology. Most of the methods described in this volume are also available and used in developing countries. The descriptions provide information about the principle of the method, possible applications, fundamentals, instruments, survey practice, processing and interpretation of the data, quality assurance, personnel, equipment, time needed, examples, as well as references and sources for further reading. Most of the remote-sensing and geophysical methods are subdivided in this way (see Table of Contents). The geological, hydrogeological, and geochemical methods are subdivided in other ways, owing to the different way they are applied in site surveys. The handbook is not intended to be used as a textbook, but, instead, to provide insights into the fundamentals, application and limits of methods, as well as case studies, selected as examples for extrapolation to other geoenvironmental concerns. This handbook cannot replace consultation with an experienced remote-sensing expert, geologist, geophysicist and/or chemist. Involvement of experts insures that the most up-to-date methods and techniques are applied. Prior starting a project, it is beneficial to first define the objectives and goals of the study so that the most suitable techniques and methods can be used. The Federal Institute for Geosciences and Natural Resources (BGR), Germany, in cooperation with scientists from universities, research institutes, and industry, has carried out two projects entitled "Methods for the Investigation and Characterization of the Ground below Waste Disposal Sites" and “Recommendations for Site Investigations of Waste Disposal Sites and Contaminated Sites in Thailand” funded by the German Federal Ministry for Education and Research (BMBF). The primary objective of these studies was to increase the understanding of the characteristics of the ground below waste disposal sites using geoscientific methods. Eight volumes of the German Handbuch zur Erkundung des Untergrundes von Deponien und Altlasten (Handbook for Investigation of the Ground below Landfills and Sites Suspected to be Hazardous) (Springer-Verlag) were prepared in the first project. The handbook “Site Investigation Methods” presented here is a product of the second project. This book contains numerous examples from both of the above-mentioned projects.
Environmental Geology, 1 Introduction Table 1-1: Geoscientific methods for site investigations: geology
7
8
1 Introduction
Table 1-2: Geoscientific methods for site investigations: landfill waste body (both operating and abandoned)
Environmental Geology, 1 Introduction Table 1-3: Geoscientific methods for site investigations: groundwater conditions
9
10
1 Introduction
The authors and editors of this volume believe that the method descriptions and case studies in this volume will illustrate the advantages of applying remote-sensing, geophysical, geological, hydrogeological, geochemical, and microbiological methods to decision-makers faced with their own environmental investigations. The consistent and knowledgeable application of methods will improve the timeliness, cost-effectiveness, and thoroughness of most environmental site assessments. Site investigation is active environmental protection. Money for effective site investigations is well invested. This is an investment in our future and that of our children. References and further reading VOIGT, H.-J., NÖLL, U., KNÖDEL, K., JENN, F., RADSCHINSKI, J., GRISSEMANN, C. & LANGE, G. (2006): Recommendations for Site Investigations of Waste Disposal Sites and Contaminated Sites in Thailand. Bangkok, Berlin, Hannover, Cottbus. (contact: [email protected]) WILKEN, H. & KNÖDEL, K. (1999): Handbuch zur Erkundung des Untergrundes von Deponien und Altlasten, Band7: Handlungsempfehlungen für die Erkundung der geologischen Barriere bei Deponien und Altlasten (Handbook for Investigation of the Ground below Landfills and Sites Suspected to be Hazardous, vol 7: Recommendations for the Investigation of Geological Barrier below Landfills and Sites Suspected to be Hazardous). Springer, Berlin.
Environmental Geology, 2 Preparatory Steps and Common Problems
11
2 Preparatory Steps and Common Problems KLAUS KNÖDEL, GERHARD LANGE & HANS-JÜRGEN VOIGT
The procedures for investigating planned, operating, or abandoned landfills, mining or industrial sites and the available methods are introduced in Part 1. The geoscientific methods are described in more detail in Parts 3 to 5. Certain preparations are required for any site investigation using ground-based methods in order to ensure the efficient implementation of field operations.
2.1 Placing of Orders and Order Handling The "rules" for commissioning site investigations differ from country to country, very often depending on national or international regulations (e.g., for World Bank or EU projects). The "rules" must be well formulated and known and accepted by both customer and the contractor. Invitations to tender must contain a specification of the investigation aims and detailed terms of references for the commissioned work. The bidding contractor has to prove that he is able to carry out the requested work. Before a bid is submitted the site should be inspected by all prospective contractors. Thus, the bidders can become acquainted with the local conditions and the customer can be questioned in detail about the proposed investigation. The situation at the site (e.g., noise sources, sealed ground surface) can hamper the application of some methods and method combinations. On the other hand, information about the terrain, accessibility and plant cover in the area to be surveyed are important for estimating the length of time needed of the field work and thus for cost calculations. Not only is information about the survey conditions important for conducting a geoscientific survey, but attention must be paid to operational safety. The opinion of an operational safety specialist may be needed to rule out any hazards for the field team. In the case of a military or abandoned ammunition site, a specialist for the removal of explosives has to be consulted. This will entail additional costs. It has to be kept in mind – that the lowest bid is not always the best one! The contract awarded to the best bidder should include a listing of the rights and duties of both customer and contractor, a description of the work to be carried out, quality assurance measures, time schedule for the work, including the delivery of the reports and data, as well as the terms of payment.
12
2 Preparatory Steps and Common Problems
The quality of the work should be checked (e.g., well logging to check the placement of screens in groundwater observation wells) by the customer. If the customer is not able to do so, the geological survey and/or the environmental protection authorities will provide advice and can help with the quality assurance of such work. Quality assurance is discussed in Chapter 2.6 and the corresponding sections in the method descriptions. The same base map (see Chapter 2.2) and the same benchmarks at the site or near it have to be used for all work carried out at a site. It is very important that the local authorities and residents be informed about the field work (see Chapter 2.3). It is also important for the investigation results to be available and transparent to the local people. Details of the rights and duties of the customer and the contractor, as well as on operational safety, are not subject of this volume. The objective here is to sensitize authorities, customers and contractors to these problems.
2.2 Collection and Use of Existing Data The most basic requirement for successful site investigations is the availability of maps and documents which facilitate a general overview of the investigation area and its surroundings and provide as much detailed information as possible. Topographic maps Topographic maps mainly display the morphology, the vegetation cover, the drainage system, and infrastructure. Scales of official maps vary between 1 : 100 000 and 1 : 10 000. A widely used scale is 1 : 50 000. Map content, however, can be rather outdated if map revisions are made only at long time intervals. While the 1 : 50 000 scale appears to be appropriate for an overview, site investigations usually require the most up-to-date maps at a considerably larger scale (1 : 10 000 to 1 : 2000). Such a large-scale map is needed as the base map for all field operations, as well as for the subsequent data documentation and interpretation. In many countries, large-scale topographic maps are not available. Therefore, base maps have to be prepared from aerial photographs or high resolution satellite images (e.g., IKONOS and QUICK BIRD). Cloud cover on the images should be less than 20 %. A ground check of the maps with a handheld GPS instrument can provide coordinates with a precision of about 10 m. This is sufficient for site investigations in most cases, however. If higher precision is necessary, maps constructed from large-scale aerial photographs (greater than 1 : 10 000) have to be geocoded and rectified by ground checks in the field (see Chapter 3.2).
Environmental Geology, 2 Preparatory Steps and Common Problems
13
Geoscientific maps Geoscientific maps are a further necessary tool for site investigations. They generally depict a variety of information about geology, hydrogeology, mineral exploration, soil, natural hazards, and land use. Although the scale of the available geoscientific maps might not always be suitable to obtain the needed information in the necessary detail, the maps nevertheless place the investigation area in its regional context and permit the survey data to be interpreted with respect to its geological, hydrogeological and structural setting. Aerial photographs and satellite images Aerial photographs and satellite images are very useful tools for planning and implementing field operations. They provide the best overview and impression of the topography and infrastructure of the investigation area. In certain cases they allow a three-dimensional evaluation. They can provide information about lineaments, geomorphological structures, and flood-prone areas, as well as land use. Aerial photographs taken at different time are helpful for reconstructing the history of a site, often with immediate relevance to environment-related problems (see Part 3).
Point data compilation An important preparatory step for a geoscientific site investigation and data interpretation is the collection of already existing information about the subsurface (desk studies). This task mainly comprises the compilation and assessment of available data from boreholes for groundwater or mineral exploration. These point data are a valuable source of information and are indispensable for the interpretation of geophysical results and calibration of hydrogeological models. Results of desk studies on the geology at the site should be verified in the field. Quarries, road cuts and other outcrops provide information about the stratigraphy, lithology and structures in the investigation area. The data should be easily accessible in an appropriately prepared and documented database (e.g., GIS).
14
2 Preparatory Steps and Common Problems
Document review For the investigation of abandoned landfills, industrial and mining sites, it is necessary to obtain the history of the site as completely as possible. For this purpose, a review of all documents on the site facilities, their construction, length of operations, as well as waste composition, quantity, and treatment methods. These documents will be normally available from the relevant authorities or the site owners. This information not only aids the interpretation of geophysical survey data and hydrogeological observations, but it is also essential for any assessment of the risk present at the site and for recommendations of remediation measures and further use of the site. As not all or even any such information may be available, it is highly useful to interview persons who lived or worked around the site during the time of landfill, mining or industrial operations.
2.3 Information Campaign and Permit Application Investigation of an operating, abandoned and planned landfill, mining or industrial site requires the approval of the relevant authorities at the state, district and/or local levels and of the affected private property owners. The objectives of the investigations, the estimated duration, and the scope of the work, as well as the methods and their likely impacts, should be explained in detail to those affected before approval for the operations is requested. Recommendations and information from public and private sources should be integrated into the working concept. As geophysical surveys and likely follow-up shallow drilling have to extend beyond the immediate investigation site, a large number of private and public landowners may be affected. Before field work is begun, permits must be applied for in order to ensure the work will be undisturbed during the site investigation. Before a permit is applied for, the customer should give the contractor a letter requesting support for the field work. The customer should also provide a site map showing the coordinates of survey lines and/or the boundary of the survey area. Depending on nature and extent of the investigation area, as well as the authorities and organizations concerned, e.g., municipal administrations, police stations and other relevant institutions must be informed on the impending field work (see check list in Table 2-1). If the investigation area includes nature, landscape, or water protection areas or other restricted areas, the governmental and/or private organizations involved must be provided information about possible impacts of the survey method(s) that will be used in order to obtain an exemption to the restrictions.
Environmental Geology, 2 Preparatory Steps and Common Problems
15
Table 2-1: Check list for the preparation of site investigations customer customer (if possible) land surveying offices, land registry offices, map distributors
district authorities municipal authorities local authorities police stations if necessary mining authorities if necessary water management authorities if necessary forestry offices if necessary road authorities if necessary railway operators if necessary occupants/leaser and/or owners of the premises
utilities operators (long-distance and local operators may be different): – freshwater – waste water – electricity (including street lighting) – gas – telecommunications specialists for the removal of unexploded ordnance (UXO), if necessary contractor
contractor contractor
provides the contractor with a letter authorizing the implementation of the work provides plans and maps provides topographic and geological maps (also as digital maps), general and detailed maps, land register maps, and names of property owners, information about topographic benchmarks to be informed and, if necessary, authorization is requested to be informed and, if necessary, authorization is requested to be informed and, if necessary, authorization is requested to be informed and, if necessary, authorization is requested to be informed and, if necessary, authorization is requested to be informed and, if necessary, authorization is requested to be informed and, if necessary, authorization is requested to be informed and, if necessary, authorization is requested to be informed and, if necessary, authorization is requested provides name, address, designation of the premises involved, private roads that may be used, authorization to use the premises, description of any damages provides information on the location of utility lines
would involve a subcontractor secures office and lodgings in the field (address, telephone and fax numbers, number of rooms, price) allocates human resources and equipment, hires local manpower if necessary provides the field team with road maps showing the way to the investigation area and the meeting point
16
2 Preparatory Steps and Common Problems
It is important to collect all available information on existing surface and subsurface installations (e.g., power lines, water mains, oil and gas pipelines, and sewer lines). If no official maps are available, local people generally know about these installations. As the locations of power lines and pipelines in an area often differ from the records provided by the operators, it is advisable to check the information on the maps with a cable/pipeline detector before starting the work. If public roads or railway tracks are crossed by survey cables it is necessary to discuss safety precautions with the road authorities or railway operators and procure written authorization. In some cases, the field investigations should be announced in local newspapers, stating time and area concerned. When the ownership of the premises has been determined, the owners or occupants must be contacted. The persons affected by the planned survey are given leaflet containing information on the site investigation and a questionnaire to obtain information about the locations of drainage ditches and other drainage facilities, private roads, power and water supply lines etc. so that appropriate measures can be taken to avoid damage. In farming areas, the contractor should meet with property owners and/or leasers to discuss the avoidance of disruptions of the field work and damage to survey equipment by agricultural machines. Immediately after termination of the survey, owners and occupants/leasers have to be contacted in order to assess any crop damage etc. Indemnity for crop damage depends on the season, on the anticipated harvest, and on the kind of damage to the land.
2.4 Mobilization and Demobilization Field work generally starts with the mobilization of a survey party or a survey group and ends with its demobilization. This involves the following work and is part of the contract: – Preparation of the survey equipment and transport to the survey site, – transport of the working party to the survey area, – establish a field office and preparation of the field work, – disassembly and loading of the equipment at the end of the field work, – transport of the working party back to the home base, and – unloading and maintenance of the equipment. The personnel and time required depends on the methods, the distance to the location, and the size of the investigation program.
Environmental Geology, 2 Preparatory Steps and Common Problems
17
2.5 Land Surveying Additional geotechnical measures, for example, drilling and cone penetration tests, or supplementary geophysical investigations are often planned within the scope of site investigations. These are carried out on the basis of the results of a geophysical survey campaign. Thus, the exact determination of measuring points and profiles is very important in order to document consistently all investigations in a site map of an appropriate scale. To properly interpret the geophysical data, it is often necessary to relate the surveyed profiles and measured points to the country's national topographic grid. The coordinates and elevation of benchmarks in or near the area to be surveyed have to be obtained from the land surveying office before the fixed points are surveyed at the investigation site. A map of the site and surroundings at a scale 1 appropriate for the aims of the survey should be made available by the customer. This will make it possible to compare the results of several contractors working in the same project area. Position determination The fixed points (benchmarks) near the investigation area must be inspected and their coordinates must be expressed in the units of the country's geodetic reference system (WGS 84, GRS 80, Universal Transverse Mercator Grid System - UTM, Gauss-Krueger system) so that the survey results can be related to the official maps. Table 2-2: The required accuracy of the geophysical measuring points and boreholes as a function of the scale of the field and presentation maps Scale of the field and presentation maps
Required accuracy
1 : 10 000
r 10 m
1 : 5 000
r 7.5 m
1 : 2 000
r5m
1 : 1 000
r2m
1:
r1m
500
The required accuracy of the geophysical measuring points and boreholes, as shown in Table 2-2, depends on the working and presentation scale. The tolerable error in the relative positions measured within the survey area depends on the applied geophysical method. It must be smaller than r 1 m. For quality assurance, the accuracy of the position and height measurements should be documented in the report.
1 The scale should be chosen so that the density of measuring points in maps and profile plots is not less than three measuring points per centimeter.
18
2 Preparatory Steps and Common Problems
If there are several benchmarks in the investigation area, the locations of the geophysical measuring points and profiles can be determined by "free positioning". The instrument (e.g., tachymeter) is positioned at a point where a line-of-sight view of the profiles to be measured geophysically and at least two, better three, fixed points is possible. The coordinates of this point can be determined from these measurements. The measurement to three or more benchmarks provides an indication of the precision of the coordinate determination for the standpoint. This method cannot be used if there is an insufficient number of fixed points. In this case, measurements must be made at several points along the profile and subsequently connected to the fixed points. In some site investigations, a local base grid is used and permanent markers placed at three base points of the grid. The locations of these base points must be marked on the maps or sketch maps included in the technical and scientific reports. The locations of the measuring points and profiles are determined with respect to this base grid using measuring tapes, an optical square and ranging poles. If possible, the local grid can be connected to the country’s grid. Altimetry The required precision for altimetry measurements is given in Table 2-3 according to the purpose they are used for. If there are no geodetic points with an elevation value near the survey area, measurements to tie the local benchmarks with remote benchmarks must be carried out. The most exact method for this objective is geometric leveling, which, however, takes a considerable amount of time, especially in rugged terrain. Measurements must be made to at least two remote points as a control on each other. This can result in long topographic leveling paths. Table 2-3: Required precision for altimetry measurements Method/object
Precision
gravity methods
r 0.03 m, for special measurements r 0.003 m
seismic methods
r 0.1 m
geoelectrics, geomagnetics
r 0.5 m
groundwater observation wells
r 0.01 m
The measurements can be made much more quickly if electronic tachymeters are used. The possibility of direct transfer of the data from the tachymeter to a computer also facilitates the data processing. The precision of the tachymeter measurements is less than that of geometric leveling, but sufficient for geophysical investigations. Because benchmarks often give only elevations or only coordinates, separate altimetry measurements are then necessary to tie the local benchmarks to the official grid.
Environmental Geology, 2 Preparatory Steps and Common Problems
19
Satellite supported positioning The American Department of Defense completed the Navigation Satellite Timing and Ranging – Global Positioning System (NAVSTAR-GPS) in 1993. This system consists of 24 operating and 3 extra satellites, with at least 4 satellites being visible from any place on the Earth at any time. Each satellite transmits modulated navigation information on two carrier frequencies (L1, L2). This information includes the orbit parameters of the satellite (ephemerides), a refraction model, a clock parameter, as well as the orbital data (in simplified form) of all satellites in the space segment. Two positioning signals (C/A code 2 , P code 3 ) are modulated on frequency L1. With the P code modulated on L2 and the navigation information, the signal traveltime from the satellite to the receiver can be determined. The P code is encoded as a safeguard against falsification (“anti-spoofing”, A-S) and is accessible only by authorized users, especially military. From the traveling time of the signals, a GPS receiver calculates its three-dimensional position (longitude, latitude and altitude). Longitude and latitude can be calculated from the data of at least three satellites and using the signals of a fourth satellite the altitude can also be determined. “Selected Availability” (SA), which reduced the attainable precision by random signal falsification, was deactivated in May 2000. When that was done, measurements could be made by nonprivileged users with greater precision. Thus, using simple GPS receivers, a precision of r 5 to 20 m can be reached for the latitude and longitude. The precision for altitude is generally lower by a factor of 1.5 to 2. Maximum precision, however, can only be attained with an optimum constellation of the satellites. How optimal the constellation is is indicated by the PDOP factors 4 of the GPS display. Multipath effects, generated at the walls of buildings, can lead to a decrease in precision. Precision can be enhanced, for example, by using differential GPS (DGPS). The error in each satellite signal is calculated from the signals received by a GPS base station with coordinates known from a non-GPS source. The correction data is then transmitted to the portable DGPS station. The precision in the GPS measurements can be within a few centimeters when both the L1 and L2 carrier frequencies are used. There are three ways to transmit this information: – Direct transmission is possible if the base station and the mobile station are only a few kilometers apart and there are no significant barriers between them. This is due to the limited range of the radio in the GPS instruments. 2 C/A code stands for Clear/Acquisition, Clear/Access or Coarse/Access 3 P code: Precise Code 4 PDOP: Position Dilution of Precision: the smaller the value the better the results that may
be expected.
20
2 Preparatory Steps and Common Problems
– As the correction signal of a reference station, however, can be used for distances up to several hundred kilometers (depending on the required precision) powerful commercial transmitters are being increasingly used for transmission of the correction data. – The third possibility is to send the correction signal via satellite. Some companies offer reference signals from geostationary satellites for a fee. Required is line-of-sight contact between the GPS instrument and the satellite. This kind of reference data transmission is of interest in nonEuropean countries. Use of satellite positioning for geoscientific purposes is a relatively new development. Besides the NAVSTAR-GPS and the Russian GLONASS systems, which can be used worldwide, the EU is installing the GALILEO system, for which more than 30 satellites will have been launched by 2008. Especially multifunctional “satellite positioning services” will increase the importance of satellite-supported positioning and navigation. These systems can be used without problems in open land, difficulties may occur in woodland or urban areas. A precision of about 1 cm to several millimeters is attainable by post-processing. The coordinates and altimetry data determined by GPS must be converted to the national or local frame of reference. Personnel, equipment, time needed How long it takes a surveying crew to execute a land survey depends on the precision required by the geoscientific methods that will be used. The survey progress is determined by topography, vegetation, visibility conditions, the distance to fixed points (benchmarks) and the survey method. Highperformance GPS systems and electro-optical tachymeters are now available, allowing precise survey data with relatively few personnel. In favorable cases, e.g., dc resistivity soundings or gravity and magnetic surveys, a land survey can be carried out by the geophysical personnel together with the data acquisition. The personnel, equipment and time needed in comparison with geophysical measurements are given in Table 2-4.
21
Environmental Geology, 2 Preparatory Steps and Common Problems
Table 2-4: Personnel, equipment and time needed in comparison with geophysical measurements Survey personnel 5 marking of points on the basis of the topography and map marking of points on the basis of the topography and map and subsequent determination of position and elevation preparation of map showing the positions and elevations of the profiles and points in the survey area
0
Equipment needed
Time 6 required [in %]
measuring tape
5 - 10
1-2
position determination by tachymeter or GPS
20 - 50
1-2
computer with appropriate software for maps
50 - 100
2.6 Quality Assurance and Reporting Quality assurance is an important part of the field work, the data processing, the interpretation, and the reporting. It must be documented in field protocols as well as summarized in the reports. Quality assurance includes x instrument checks before, at certain time intervals during, and after the measurements; x careful maintenance and calibration of equipment; x entry of all relevant occurrences and possible noise sources (e.g., passing cars during the geophysical measurements, heavy rain, storm, steel fences, power lines) in the field protocols; x daily data check and plotting of results during the field campaign; x repetition of a certain percentage of the measurements in order to estimate measurement errors; x documentation of all steps and parameter values used in the data processing and interpretation. x The report must contain all information needed to understand the report. x The figures and maps must be clear, understandable, and convincing.
5 Land survey personnel in addition to the personnel needed for the geophysical survey. 6 Time needed for the land survey compared to the time needed for the geophysical survey.
100 % means that the same time is needed for the land survey as for the geophysical measurement.
22
2 Preparatory Steps and Common Problems
x The conclusions of the report must answer the question whether and how far the tasks and targets were achieved, must list remaining problems, and must be understandable for non-experts. x All measured data, relevant intermediate results, as well as figures and maps of the final report must be submitted to the customer in well documented form in both hardcopy and machine-readable form (e.g., CD or floppy disk) for later use as evidence in possible legal dispute and/or reinterpretation together with results of other investigations and, if necessary, with supplementary measurements. For the specific aspects of quality assurance of methods, see the “Quality Assurance” sections in Parts 3 to 5.
Environmental Geology, 3 Remote Sensing
23
3 Remote Sensing PETER HERMS, BERNHARD HÖRIG, FRIEDRICH KÜHN, DIETMAR SCHMIDT & ANKE STEINBACH, with a contribution by TILMANN BUCHER
3.1 Aerial Photography DIETMAR SCHMIDT & FRIEDRICH KÜHN
3.1.1 Principle of the Methods Despite all technical progress in digital imaging, interpretations of standard aerial photographic images remain an important remote-sensing tool. Aerial photography can address a multitude of geoscientific questions and can be highly effective when used for logistics and planning (BÖKER & KÜHN, 1992). The cost of aerial photography is rather low. The data are informative, easy to manage, and the film does not require special image processing resources for analysis. Cameras for aerial photography, also referred to as metric cameras and frame reconnaissance cameras, are analog to standard photographic cameras using lenses, shutters, and film. These cameras, however, are more expensive, and mechanically and electronically complicated since they need to adjust the orientation of the camera during data acquisition. Ideally, aerial photography equipment is coupled to a Global Positioning System (GPS) to ensure better precision and accuracy. Photographic multispectral cameras have been widely used in planes and satellites in the 1970s and 1980s (COLWELL, 1983; KÜHN & OLEIKIEWITZ, 1983). Since then, the application of such cameras has decreased for several reasons: their complicated operation mode, spectral distortion depending on the angle of incidence on the camera lenses and filters, and the complicated and imprecise evaluation procedures. They have been replaced by multispectral digital imaging systems (see Chapter 3.3). Although there are a variety of types of film available for aerial photography, three kinds are used in almost all situations. These commonly utilized films are panchromatic (black and white), color and color-infrared
24
3.1 Aerial Photography
(CIR) film. The latter one is able to serve most environmental and geological tasks. Vertical aerial photographs provide a three-dimensional impression of an area. Interpretations of stereo-pair photographs require an overlap of 60 - 90 % between two images along the flight line. An overlap of approximately 35 % is needed between adjacent flight lines to allow for flight irregularities. As the distance between any two consecutive images is greater than the distance between the human eyes, the topography of the target area appears strongly enhanced and exaggerated when the images are viewed stereoscopically. This enhancement commonly provides detailed information for geological and environmental analysis. Aerial photographs can be used for, but are not limited, for detection of man-made features. In addition, aerial photographs can often be analyzed to identify rock formations and types of soil, typical relief forms, distinctive vegetation types, drainage patterns and specific types of land use via changes in color or gray scale. In general, as the frequency of man-made features increases in an area, it becomes more difficult to extract geologic information from aerial photographs. Even older aerial photographs provide proper spatial resolution and often document changes in an area of interest over a period of several decades. They allow the evaluation of geological and initial environmental situation. Archives of aerial photographs are maintained worldwide, e.g., in governmental survey offices, and can provide aerial photographs at low cost. A thematic interpretation of a specific site using aerial photographs is usually carried out at scales between 1 : 2000 and 1 : 10 000. A topographic base map at the selected scale is necessary. The preparation of topographic maps is described in Chapter 3.2. Regional analyses (e.g., lineament analyses) are carried out at scales between 1 : 25 000 and 1 : 100 000.
3.1.2 Applications x Chronological analysis of the development of land use, waste disposal, mining, and industrial sites and their surroundings, x assessment of the previous geological and environmental situation at sites now covered by waste disposal, mining, and industrial facilities, x search for seepage of water at the edges of landfills and mining waste heaps, x locating springs and moisture anomalies, x investigation of natural and artificial drainage systems, x locating areas of high and low permeability to water,
Environmental Geology, 3 Remote Sensing
25
x recognition and monitoring of areas of destabilization, x investigation of vegetation vitality, x assessment of surface water conditions, x delineation of fractures and lineaments, x inventory of sites suspected to be contaminated, x mapping of land use patterns and biotops, and x search for potential new waste disposal sites.
3.1.3 Fundamentals Remote sensing utilizes electromagnetic (EM) radiation in the ultraviolet, visible, infrared, and microwave portions of the EM spectrum to obtain information about the Earth’s surface with sensors carried by either an aircraft or satellite. Active and passive methods are used for remote sensing. Passive methods (see Chapters 3.1 - 3.3) use reflected solar radiation and radiation emitted from a surface. Active remote-sensing systems (Chapters 3.2 and 3.3) have their own source of radiation (e.g., radar or laser). Interaction of electromagnetic radiation with the Earth’s surface provides information about the reflecting or absorbing materials. Due to the small wavelengths of EM radiation used (nanometer to micrometer range) there is limited penetration of the targeted (optically thick) land surface objects. Given the proper atmospheric transmission, remotely sensed data represent the Earth surface conditions. Because EM radiation is the carrier of the remote-sensing information, it is important to define the different ranges of the EM spectrum. In the literature, it is difficult to find an unambiguous division of the electromagnetic spectrum, since it is of continuous nature. Divisions are commonly based on type of sensor and response of natural materials, etc. We have decided to use the divisions of the spectrum of ERB (1989). In addition to ERB’s divisions, we will use a further subdivision of the middle infrared (MIR) into MIR-I and MIR-II (Table 3.1-1) for remote-sensing purposes. The sun is the primary source of the Earth’s incident EM radiation. The energy spectrum of the sun is almost identical to that of a black body at 5900 K. Solar radiation reflected from the Earth’s surface is in the NUV to NIR-II (0.315 - 3.0 µm). However, there are gaps in this spectral range due to scattering, absorption, and reflection by gases, particles, and other atmospheric constituents. Major atmospheric absorption bands appear at 1.4 µm, 1.9 µm, and 2.5 - 3.0 µm resulting from atmospheric water vapor and carbon dioxide (KRONBERG, 1985; KUEHN et al., 2000). Little or no solar radiation reaches the Earth’s surface at these wavelengths and they cannot be used for land remotesensing purposes.
26
3.1 Aerial Photography
Table 3.1-1: Spectral range of electromagnetic radiation utilized by remote-sensing sensors; spectral divisions by ERB (1989) Radiation near ultraviolet visible light near infrared middle infrared (thermal infrared) far infrared microwave (radar)
Abbreviation NUV VIS NIR-I NIR-II MIR MIR-I MIR-II FIR MW
Wavelength O [in µm] 0.315 0.38 0.78 1.4 3.0 3.0 8.0 50.0 1000
– 0.38 – 0.78 – 1.4 – 3.0 – 50.0 – 5.5 – 15.0 – 1000 – 106
Solar radiation is either transmitted, absorbed or reflected at the surface. Reflections can be either direct or diffuse. The proportions of reflection, absorption, and penetration of the incident radiation at the Earth’s surface depend on the physical, chemical, structural, and textural properties of the surface. Molecules and chemical bonds (which make up all materials, including soils, rocks, water, and plants) are characterized by a specific energy level. The absorption of the incoming radiation provides the energy required for the transition from one energy state to another. Remote-sensing systems record the reflected radiation that reaches the sensor after its interaction with the Earth’s surface. Hence, the properties of the soil, rock, and other materials at the Earth’s surface are indirectly recorded in the data or images. The maximum energy of solar radiation is within the visible part of the spectrum (VIS) at 0.48 µm. The maximum radiation from the Earth is at about 9.7 µm. Thus, the best range for recording thermal radiation from the Earth’s surface is between 8 and 12 µm. Within this spectral range, slight temperature variations resulting from differences in the type of soil, soil moisture, or the presence of specific pollutants can be identified. Therefore, thermal scanners (Chapter 3.3) are designed to be most sensitive in the spectral range of 8 - 12 µm. To investigate hot surfaces (e.g., radiation from a blast furnace or the structure of a lava flow), a spectral range of 3 - 5 µm is more suitable. Radiation in the visible and infrared portions of the spectrum is absorbed by certain constituents of the atmosphere. Major absorption bands lie between 5 and 8 µm and between 13 and 30 µm. These spectral regions are usually not used for land remote sensing. Applications of thermal remote sensing are described in Chapter 3.3. Vertical aerial photographs are commonly produced at scales of 1 : 20 000 (sometimes 1 : 50 000) to 1 : 5000 for applications in environmental geology. Using a lens with a focal length of 150 mm, the flight altitude required to
Environmental Geology, 3 Remote Sensing
27
produce these scales is between 3000 m and 750 m above ground, respectively. Standard film size of 23 cm u 23 cm then will capture an area of 4600 m u 4600 m (at a scale 1 : 20 000) or 1150 m u 1150 m (scale 1 : 5000). The basic parameters of aerial photography, i.e., the scale (S) of the aerial photograph, the field of view (A) of the camera, and the ground resolution (Rg), can be calculated using the following equations (BORMANN, 1981a): S
A
Rg
cf hg hg k cf
hg k RL
where c f Sn hg k RL
1 , Sn
,
(3.1.1)
(3.1.2)
,
(3.1.3) is the focal length of the camera lens, scale constant, flight altitude above ground, film frame size (usually 23cm u 23 cm), and line resolution (lpi) of the film according to the manufacturer.
Equation 3.1.3 gives an estimate of the spatial resolution, Rg. To precisely calculate the resolution of a given “film–lens” system, a contrast modulation function has to be used which describes the relation between the degree of contrast in the terrain and in the image as a function of object size for a given spatial frequency domain (BORMANN, 1981b).
3.1.4 Instruments and Film Aerial photographs are used for both thematic interpretation and photogrammetry. The equipment and quality standards are mainly the same. They differ only in the films used. The following equipment and materials are required for aerial photo surveys: x Large format aerial cameras (Table 3.1-2) are available with different focal lengths (Table 3.1-3). Modern survey cameras (Table 3.1-2) are fully equipped for movement compensation (IMC – Image Motion Compensation). Blurring caused by the forward movement of the aircraft, can be compensated by FMC (Forward Motion Compensation). Fuzzy film images resulting from vibrations of the aircraft are prevented by using a gyroscope mount. x The “Flight Management System” (corresponds to IGI’s CCNS-4, T-FLITE of ZI/IMAGING or Ascot of LH-Systems), which was specially developed for survey flights on the basis of GPS, enables higher precision
28
3.1 Aerial Photography
in flight management and more accurate shutter release for the specified areas. x DGPS (Differential Global Positioning System, see Chapter 2.5) receiver in the aircraft and on the ground, as well as GPS for terrestrial surveying. x Suitable film (see below in this section). x Equipment for the developing of black and white, color and/or CIR aerial film. Table 3.1-2: Technical specifications for a large format aerial camera
manufacturer format film magazine capacity exposure cycle time weight
ZEISS RMK TOP
LEICA RC 30
ZI/Imaging 23 cm × 23 cm 152 m 1.5 s ca. 150 kg
LH – Systems 23 cm × 23 cm 152 m 1.5 s ca. 150 kg
Table 3.1-3: Lens types for aerial cameras, ALBERTZ (2001) Lens Type narrow angle normal angle intermediate angle wide angle super wide angle
Focal length [cm] 61 30 21 15 9
Maximum angle 33 gon (30 degrees) 62 gon (56 degrees) 83 gon (75 degrees) 104 gon (94 degrees) 134 gon (122 degrees)
Depending on the task and the desired results, different kinds of film can be used for aerial photography: Panchromatic black-and-white film In many countries the most common film used for aerial photographs is black and white panchromatic film. This film is frequently applied for geodetic and cartographic purposes, but may also be used for specific thematic mapping tasks. The high spatial resolution allows for identification of minor fractures. Phenomena such as changes in rock and soil type, soil moisture, or secondary effects related to contamination can also be detected with this film. Stereoscopic techniques using pairs of aerial photographs allow an experienced interpreter to characterize objects at the surface of a site in detail. SCHNEIDER (1974), KRONBERG (1984) and CICIARELLI (1991) show a multitude of applications for panchromatic film. Updates of topographic maps are in many countries made with panchromatic black-and-white film (Fig. 3.1-1).
Environmental Geology, 3 Remote Sensing
29
Fig. 3.1-1: Black-and-white aerial photograph taken by the Royal Thai Survey Department on November 11, 1995, showing the more than 35 m deep Nong Harn borrow pit near Chiang Mai in northern Thailand. The pit is being filled with waste from the west along a road to the bottom of the pit. The pit has steep slopes and was previously about 45 - 50 m deep. The western part of the pit floor is covered by more than 10 m of waste. The southern, western, and eastern slopes are covered by a plastic liner. The fine white lines on these slopes are ropes to hold down the strips of the liners. The northern slopes have the ropes too, but no plastic liner cover, as indicated by the brighter gray color and by the micro-topography of the uncovered slope.
Infrared film Black-and-white film can be sensitized to extend beyond the visible range of light (VIS) into near-infrared wavelengths (NIR-I). This part of the nearinfrared spectrum, also referred to as the photographic infrared, ranges from wavelengths of 0.78 µm to about 0.9 µm. EM radiation detected by nearinfrared film is reflected solar radiation and should not be mistaken with thermal infrared radiation. The latter is the EM range between 8 and 15 µm and is used for thermal imaging. Due to the intense absorption of infrared radiation by water, near-infrared film is highly sensitive to changes in soil moisture content. Also, near-infrared film can be used to detect changes,
30
3.1 Aerial Photography
extent, and state of vegetation cover. Healthy vegetation intensely reflects radiation in the same range. Color infrared film (CIR) has largely replaced black-and-white infrared film for investigations of damage to forests and other environmental conditions. Color and CIR film Color aerial film is used to depict a terrain in natural colors. Color film has three light-sensitive layers for the visible wavelengths of blue, green, and red. By substituting a near-infrared-sensitive layer for the blue-sensitive layer, a false-color photograph can be produced. This near-infrared-sensitive layer reacts to the intense reflection of NIR-I radiation by vegetation. As a result, if a reversal process is used for developing the film, the chlorophyll-rich vegetation will be represented by intensive red colors and low chlorophyll content by pale grayish red. Because there is less scattering of infrared radiation in the atmosphere than that in the visible range, CIR film (Fig. 3.1-2) yields sharper images with greater contrast from high altitudes than normal color film. The increase in contrast is enhanced by the use of a yellow filter to suppress the blue parts of light. CIR film has proven indispensable for a broad range of thematic mapping objectives. Environmental mapping relies on CIR aerial photography because of its high sensitivity to changes in plant type and vitality. Methods for mapping the vitality of trees in the vicinity of a waste disposal site using CIR images are described in Section 3.1.6.
Environmental Geology, 3 Remote Sensing
31
Fig. 3.1-2: CIR photograph, taken by Hansa Luftbild (German Air Survey) together with Sky Eyes (Thailand) on December 22, 2000, showing the refilled Nong Harn borrow pit near Chiang Mai in northern Thailand. The site is surrounded by a forest of several species of trees, recognizable by different colors and shapes of the crowns. Compaction of the 45 - 50 m thick waste in the pit creates a depression cone, visible by the blue round pond in the middle. Usually, water in CIR photographs appears dark or black. The blue color indicates relatively high turbidity. Chemical reactions within a landfill produce gas, which carries water with it as it ascends and escapes at the surface. The red arrows show the flow direction of surface water towards the pond at the surface.
Archival aerial photographs Generally, any characterization of hazardous sites starts with an archival search for aerial photographs to reconstruct the historical development of the site and its surroundings. These photographs are commonly panchromatic black-and-white. Archives of aerial photographs are maintained by governmental survey offices, municipal administrative offices, commercial aerial photography firms, military institutions, private archives, and users of aerial photographs.
32
3.1 Aerial Photography
Fig. 3.1-3: World War II aerial photograph (March 24, 1945) of a destroyed industrial complex in the Dortmund area (Ruhr district, Germany). Damaged industrial plants, from which pollutants may have seeped into the soil, and bomb craters and pits, which may have served as dumps, are potential hazardous sites (courtesy of Luftbild-Datenbank, Ing.-Büro Dr. Carls, Estenfeld).
Most wartime aerial photographs were taken during World War II. More rarely, photographs from times before are available as well. Target Information Sheets exist for most wartime reconnaissance flights; these define the rationale for and the outcome of each bombing. In general, it is possible to use these sheets to deduce which potentially toxic substances seeped into the soil after bombing an ordnance plant or a chemical factory. These sites should be considered to be potentially hazardous. Although aerial photographs taken
Environmental Geology, 3 Remote Sensing
33
as part of a land survey usually cover large areas, military photographs are generally taken on linear traces with sudden direction changes (DECH et al., 1991). The archives of the American and British governments are additional sources of historical aerial photographs. There are also archives of aerial photographs which were taken during military actions by the American or British army since World War II. Figure 3.1-3 shows an archival aerial photograph taken in 1945. In general, private consulting companies are specialized in archive searches for these wartime aerial photographs. Oblique aerial photographs Geo-environmental assessments may require oblique aerial photographs, when overviews of large areas are needed (Fig. 3.1-4). Oblique photographs may prove particularly valuable in the early phases of a project when the overall site is being characterized.
Fig. 3.1-4: Oblique aerial photograph taken on May 11, 1993, looking westward across the waste incineration facilities at Gallun, south of Berlin, Germany, (foreground) and the northern part of the Schöneiche landfill (middle ground) to the Schöneicher Plan landfill in the background. (Photo: F. BÖKER and F. KÜHN, BGR).
34
3.1 Aerial Photography
Interpretation instruments The thematic evaluation and interpretation of aerial photographs can be carried out using a mirror stereoscope or a digital stereoplotter. Stereoscopes are available in various shapes and sizes.
3.1.5 Survey Practice Important aspects in planning and carrying out an aerial photography survey are selection of the camera, objective and navigation system, scale, ground resolution and flight altitude, flight path and spacing of flight lines, selection of a suitable film, choice of season and time of day, weather conditions, and ground check and logistics. For scale, ground resolution and flight altitude see Section 3.1.3. State-of-the-art instruments and films are described in Section 3.1.4. Aerial photography surveys, which now are semi-automated with modern navigation and camera systems, require extensive sophisticated equipment and techniques. As already mentioned in Section 3.1.1, aerial photographs should be taken with 60 - 90 % overlap in the direction of flight for stereoscopic restitution to be carried out. An overlap of approximately 35 % is needed between adjacent flight lines to allow for flight irregularities (Fig. 3.1-5). Current aerial film has different spectral sensitivities and geometric resolution (see Section 3.1.4). The film must be transported from the flight area to the photo lab and developed strictly following manufacturer’s instructions. Ground control point coordinates for rectification of aerial photographs may be determined after a flight and the film has been developed in a separate GPS survey if no reference data set is at hand. For this task it is necessary to select and carefully mark control points in the aerial photographs as well as some additional points that can be used if the selected ones are inaccessible on the ground. During the ground survey it is helpful to carry out photographic documentation or sketches of the terrain at the control point to confirm the relationships between the images at later stages. Coordinates of ground control points must always be determined at ground level in order to avoid a shift between the on-site measurement and the control points on the aerial photograph (e.g., roofs and masts). Particularly suited for control points are geometric man-made features such as street intersection and trail crossings, junctions, bridges, and the corners of buildings and walls.
Environmental Geology, 3 Remote Sensing
35
Fig. 3.1-5: A 60 % forward overlap and a 35 % flight-path overlap (sidelap) are necessary for effective stereoscopic evaluation (3-D) of photo pairs, after SCHNEIDER (1974)
The time of acquisition of the aerial photography is a critical factor in maximizing the value of aerial photographs for interpretation (BÖKER & KÜHN, 1992). Vegetation often covers geological features. Therefore, photographs for geologic interpretation should be taken during seasons with minimum vegetation. In temperate climate zones for example, aerial photographs taken in early spring after the snow has melted and the surface has dried up, and before the vegetation has fully emerged, are well suited for separating different kinds of soils and rocks and for detecting fractures and permeability anomalies. On the other hand, the existence of pollutants in specific areas may be recognized on the basis of certain vegetation patterns. In this case, and for mapping land use patterns and biotopes, it is better to use aerial photographs taken during the summer. Data about damages to vegetation caused by contamination should be acquired during the second half of the growing season in the late summer to early autumn. Taking photographs during this season avoids both the springtime vegetation burst, when small or moderate changes in vitality are hard to detect, and the rapid seasonal decay in late autumn. In the climates of the subtropics and tropics, the alternation between the temperature and moisture conditions of rain and dry seasons determines the time of aerial photo acquisition. For investigations of barren soils, rocks, and geological structures the best time is at the end of the dry season, when the vegetation cover is at its minimum. If the vegetation itself is the object of
36
3.1 Aerial Photography
evaluation, e.g., for land use or biotope mapping, forest inventory and vitality evaluation, the beginning of the dry season is the best time for the photographic survey. The vitality of the vegetation is commonly still high at this time and visibility from the air is good for taking aerial photographs. Weather conditions are an important factor when acquiring images for geologic interpretation because soil moisture can greatly affect the quality of data interpretations. In some instances, it will be even useful to acquire data at a low altitude of the sun (10° to 30° above horizon), since low-relief forms appear enhanced by the more pronounced shadows on the ground. This increases the chance of detecting subtle geologic features, such as fractures. Consequently, the experience of the remote-sensing specialist is crucial in determining the appropriate strategy for acquisition and interpretation of aerial photographs. The interpretation should be combined with at least two ground checks, one just before the survey – to establish a classification scheme, i.e., interpretation key – and one at the end of the survey.
3.1.6 Interpretation of Aerial Photographs The most important image characteristics and landscape factors used in mapping vegetation, soil, rocks, and geologic structures from aerial photographs are: x photographic gray scales, x morphology, x vegetation, x drainage systems, x geologic structures, and x patterns related to land use.
Fig. 3.1-6 (next page): Generalized schematic of a landfill with a permeable base and without a drainage water collection system (top is a plan view and bottom is a cross section) showing structures, features, and properties that may be identified with remotesensing techniques, depending on the site conditions (examples in Section 3.1.9)
Environmental Geology, 3 Remote Sensing
37
38
3.1 Aerial Photography
These features are described, discussed and illustrated by MILLER & MILLER (1961), KRONBERG (1984) and SABINS (1996). An evaluation of the environment using remote-sensing data requires consideration of further aspects (example in Fig. 3.1-6). Indicators of tectonic fractures, potential pathways for pollutant migration, and the water retention potential of soil layers, as well as anthropogenic features of a landscape, need to be considered. The most practical features for identifying anthropogenic components of a landscape in aerial photographs include the following (DODT et al., 1987): x anomalous topography that may indicate waste disposal sites, mine spoil heaps, excavation work, etc., x anomalous gray-scale patterns that might indicate contaminated soils, filled-in depressions, spread of leachate from waste disposal sites, abandoned industrial sites, etc., and x vegetation anomalies that might indicate contaminated soil and groundwater. Conventional and digital processing methods are used for the interpretation of aerial photographs with digital processing having become more important in the last years. Digital mapping and interpretation systems use the original aerial photographs, which are interpreted using a high quality stereoscope with a zoom lens, as well as computer hardware and software to calculate image coordinates and parallax for precise determination of distances and terrain elevations. Such systems allow mapping and GIS-related recording of terrain features. Photogrammetric mapping systems use a pair of stereoscopic orthophotographs on the screen, where they can be viewed and interpreted through a special pair of glasses. It has to be evaluated before starting a project, whether the objectives of the interpretation can be achieved with the available aerial photography. Generally, detrimental impacts on water, soil and vegetation can be better understood and evaluated when previous events or the initial conditions are known. Therefore, if human activity at the site occurred over a long period of time, historical photographs of the site and its surroundings should be incorporated as much as possible. Color or better CIR aerial photographs are especially suitable for such a site investigation. Chronological development of waste disposal, industrial or mining sites In order to assess the contamination potential of waste disposal, industrial or mining sites multi-temporal analyses of archival aerial photographs (Figs. 3.1-7 and 3.1-8) have been proven to be a useful tool. This method
Environmental Geology, 3 Remote Sensing
39
requires aerial photographs taken frequently over a relatively long period of time to investigate spatial and temporal changes in natural and man-made features. Historical photographs reveal the kind of activities at the site, where and how the waste disposal started, the type of waste (liquid, scrap, dust, etc.), and disposal of dangerous waste in barrels or as mud. These investigations are even more important when no documents or only fragmented records about the activities exist. Multi-temporal analyses of archival aerial photographs are also useful for an inventory of sites suspected to be hazardous. The results can be stored and updated in a geographical information system (GIS – see Part 6). Evaluation of the former environmental and geological situation at the location of a current landfill Many old waste deposits are in natural or excavated holes, such as sinkholes, caves with openings at the surface, abandoned quarries, and clay or gravel pits. Remote-sensing data predating a landfill can be used to estimate its present subsurface characteristics (Figs. 3.1-9 and 3.1-10). Under certain conditions, current remote-sensing data of the surroundings might be extrapolated to zones below the landfill, but archived aerial photographs are normally the best source of such information. Shallow geophysical methods can also be used to estimate the subsurface characteristics, but at greater expense and more limited capability to characterize the layers below the landfill. For the investigation of such sites the following aspects are relevant: x surface characteristics prior to landfill operations, x spatial distribution of impermeable and permeable soil and rock (to determine potential areas for spreading of leachate), x formerly waterlogged areas now underlying present landfills, x natural and/or artificial drainage systems that provide pathways from the waste disposal site into the surrounding area. Springs, permeability and soil moisture anomalies, indications of water seepage, and natural and artificial drainage systems Springs and moist areas associated with a lack of ground cover usually appear darker in black-and-white aerial photographs. In CIR images, areas of high moisture content can also be recognized by the associated vegetation shown as intense reddish colors. Signs of wetness on the surface may be associated with slight changes in topography or with the presence of impermeable, nearsurface beds. Anomalous plant growth, discoloration of soil, or visibly wet areas at a landfill’s edges indicate possibly contaminated seepage water and potential groundwater contamination (Fig. 3.1-11). Plant growth depends, among other factors, on the permeability of the soil and the underlying beds.
40
3.1 Aerial Photography
The lighter colored areas in Figure 3.1-12 correlate with areas of poor plant growth due to the sandy soil and the permeable beds there. Precipitation and leachate from the neighboring landfill can migrate through this permeable material directly into the aquifer. If wet areas are detected near to a waste disposal site, as shown in Figure 3.1-13, chemical analysis of surface and groundwater samples is recommended. Natural surface and subsurface “channels”, like frost cracks in glacial till and erosion trenches in boulder clay filled with sandy, permeable material can act as pathways for leachate from landfills, industrial and mining sites. An artificial drainage system and drainage ditches may have to be installed when the water flow is blocked by impermeable rocks and soils or when the water table is high (Fig. 3.1-13). During deposition of the waste, drainage ditches are often filled with permeable materials, such as excavated soil, building rubble, or waste. In contrast, irrigation ditches have to provide, for example, agricultural fields with additional water during dry seasons or when the water table is too deep for the plant roots to reach it. A dense network of irrigation ditches through cultivated land, in rice paddies for example, may later be covered by a landfill (Fig. 3.1-14). Waste disposal and other sites suspected to be hazardous have to be investigated for signs of artificial or natural drainage systems. The results of these investigations help to determine the pathways for the migration of contaminated leachate and identify potential hazards. Recognition and monitoring of destabilization of the ground Knowledge about the stability of the ground is necessary for risk assessments of waste disposal, industrial and mining sites. Recent and historical aerial photographs often contain indications for destabilization of the ground, e.g., subsidence caused by subrosion and mining, landslides or sinkholes. Terrain features indicating destabilization of the ground are fracturing, subsidence of the land surface, caving to the surface, slippage along bedding planes on slopes, breaks, as well as tensions and relaxation indications. These features can also be seen in the aerial photographs as changes in topography, soil moisture, roughness, water permeability and vegetation. If the slope angle of landfill or mining spoil heaps and/or other geotechnical parameters are not suitable, collapses of the deposited material can be expected. Such features can also be observed and monitored by aerial photography (Fig. 3.1-15). Vitality of vegetation Plants, in particular trees are very good indicators of the presence of hazardous substances in the air, soil, and groundwater if they are in contact with the contamination via roots or leaves. CIR film is a good tool for evaluating such damage because of its sensitivity to the near infrared (NIR-I) wavelengths. This part of the electromagnetic spectrum indicates the condition of vegetation
Environmental Geology, 3 Remote Sensing
41
that is not observable to the human eye. The appearance of a tree under environmental stress is different from that of a healthy tree. The red color of healthy vegetation in CIR aerial photographs results from the strong reflection of near-infrared radiation by the chlorophyll in leaves. In general, lower chlorophyll content, causes less reflection of NIR-I radiation. Thus, the color of unhealthy leaves and needles in the photographs will become less red in proportion to the degree of vegetation “greenness”. Different tree species often show different color changes in CIR photographs in response to stress, e.g., damaged oaks and willows turn black, birches, alders, and willows turn brown, linden trees (Tilia) turn bluish-green, and silver willows become whiter. In addition, a decrease in tree vitality is also recognizable by the typical thinning of the crown, caused by the loss of leaves and small branches. In the photograph, a decrease in vitality turns a crown into a more open shape with visible branches. A vitality scale with 4 to 5 steps can be worked out for each species on the basis of changes in the color, interior structure and shape of the crown. These three criteria were developed during forest inventories using appropriate interpretation keys (MURTHA, 1972) and have also used in tropical forests (DE MILDE & SAYN-WITTGENSTEIN, 1973; TIWARI, 1975; MYERS, 1978). Such forest surveys in Thailand using aerial photographs have been described by TANHAN (1989). A color key for photo interpretation developed immediately after the aerial CIR photographs are taken can eliminate potential errors due to seasonal or weather-related changes in vitality. Areas with a shallow groundwater table, where the transport of the pollutants and contact with the roots is ensured for a long period of time and over a large area, are best suited for detection of contamination. In contrast, if a waste site is situated on top of a hill, where there is a deep groundwater table, the method might not work. Wind causing movement of the vegetation at the time the aerial photographs were taken reduces the applicability of the method. The following issues must be considered for an assessment of vitality: x only trees of the same species and similar age are comparable and x local and biological characteristics for differences in vitality (e.g., local dryness or moisture, parasites, diseases) should be excluded. This is especially important in regions with a distinct separation of wet and dry seasons. Even during the dry season, changes in topography and associated groundwater levels, springs or different permeability of underlying deposits play an important role. Except for the evaluation of isolated trees, the analysis of a wood’s edge may reveal trees which have been more intensively damaged by frequent exposure to toxic substances transported by wind than the trees deep inside the forest. Thus, by including information about prevailing wind direction, the toxic emissions may be traced back to their origin. Suitable scale to evaluate trees in a CIR photograph is between 1 : 2000 and 1 : 10 000. Because shadows can disturb the interpretation, the sun should
42
3.1 Aerial Photography
be high above the horizon during photo acquisition. Figures 3.1-16 and 3.1-17 show the influence of contaminated water seeping from a landfill towards nearby trees in an area with a high groundwater level. Assessment of surface water conditions Surface water and groundwater from waste disposal, industrial, mining or agricultural sites may contain contaminants. Such contamination in receiving streams, ponds and lakes is indicated in CIR aerial photographs as cloudiness and changes in the color of surface water. Water bodies are normally black in CIR photographs since water absorbs most of the radiation. In this context, color photographs including shorter wavelengths of the visible blue and green are more suitable since they are less absorbed and penetrate deeper into the water surface. The Fig 3.1-18 photograph indicates blue colors for water bodies emphasizing the stronger reflection in shorter wavelengths due to higher concentrations of insoluble contaminants and algae. Lineaments Fractures are often easily detected in aerial photographs (Fig. 3.1-19). They are recognizable as linear changes in topography, drainage patterns or patterns of vegetation, as well as by different colors of rocks and soils (MILLER & MILLER, 1961; KRONBERG, 1984 and SABINS, 1996). If morphologically expressed, fracture systems can be best observed in images obtained at low sun angles. Additionally, fractured rocks show signs of enhanced weathering and erosion. Based on information about fractures obtained from the evaluation of aerial photographs, further geophysical investigations can be assisted and better planned. Biotope and land use mapping Aerial photographs are a primary source for the mapping of biotopes and land use, because of their excellent spatial detail they provide. Because different kinds of plants can be better distinguished in CIR photographs than in normal color or even panchromatic aerial photographs, they are essential for this kind of problem. Such surveys are carried out worldwide for planning and regional natural resource management. Figures 3.1-20 and 3.1-21 show an example of a poorly chosen location of an industrial plant and its disposal site immediately next to Esch-sur-Alzette in Luxembourg. The legend and key for interpretation depend on the legal requirements, standards for regional planning documents and the pattern of land use. The standard system for land use mapping in Thailand was designed by the Department of Land Development in 1982, based on black and white aerial photographs at a scale of 1:15 000 (WACHARAKITTI, 1982).
Environmental Geology, 3 Remote Sensing
43
The mapping of biotopes differs from land use mapping. The botanical and forestry classification is more detailed. The quality of such an inventory depends on the expertise of the remote-sensing specialist to recognize all categories of the rather sophisticated classification system. Nevertheless, the interpretation results need to be confirmed by ground checks. The time required for interpretation depends on the complexity of the classification scheme. Classification systems for biotopes with hundreds of different types are common in Central Europe. Search for new waste disposal sites Remote-sensing data, particularly when combined with more traditional geologic and geophysical tools, can be used during the planning and development of new disposal sites. Information on the following features can be derived from aerial photographs: a) impermeable rocks at the location of the planned waste disposal site, b) existence of natural and artificial drainage systems, c) substrate with a high permeability, d) fractures, e) shallow groundwater table at the site of interest, springs and marshy areas, f) indications of landslides, and g) previous human activities (e.g., former gravel pits, quarries, mining) which might have increased permeability of the underlying rocks.
3.1.7 Quality Assurance For high quality aerial surveys in site investigation, the basic technical standards must be met (SCHWEBEL, 2001). In Germany, these standards are obligatory in accordance with the DIN standard (DIN 18740 - 1) and European standards at a similar level are in preparation (BALTASAVIAS & KAESER, 1999). The quality of site investigations by aerial photography depends on: x the equipment and the technology used, x the conditions when the aerial photographs are being taken, and x the quality of the film processing and photo interpretation.
44
3.1 Aerial Photography
Equipment and techniques: 1. Large format aerial cameras must be recalibrated every two years in accordance with ISPRS guidelines (ISPRS = International Society of Photogrammetry and Remote Sensing). 2. The camera should be mounted on a gyroscope. 3. The focal length of the camera lens must be selected on the basis of the topography in the area of investigation. As a rule, the focal length should not exceed that of a wide angle lens. 4. Aerial photographs should be taken with full IMC (Image Motion Compensation) and FMC (Forward Motion Compensation). 5. High-precision DGPS must be used based on ground reference stations closer than 50 km from the project area. 6. High-resolution aerial film that yields at least 30 lp/mm (lp = visual line pair) must be used. 7. Storage and usage of film: Experience is needed if good results are to be obtained with film for aerial photography, especially CIR film. The film’s age and how it is stored before and after exposure influence the quality of the photographs. The film should be stored according to the manufacturer’s instructions (e.g., at -18 °C). A film stored at very low temperatures should be defrosted before use and should be kept cold during transport to the photo laboratory. The film must be developed following the manufacturer’s instructions. For quality assurance, sensitive CIR film should be handled according to the VDI guidelines (VDI-GUIDELINE 3793, 1990). 8. Appropriate filters must be used to obtain images with optimum color differentiation. This means a blue filter is required for CIR film. 9. Photographic scales between 1 : 2000 and 1 : 10 000 are recommended. 10. Forward overlap of 60 % and a sidelap of 35 % are necessary for stereoscopic processing (Fig. 3.1-5). 11. It is important that the maximum deviation from the planned flight line does not exceed 100 - 150 m. 12. The preferred flight direction is north–south. Conditions while taking aerial photographs: 1. A cloudless sky and a minimum visibility of about 10 km are required. There should not be any ground haze.
Environmental Geology, 3 Remote Sensing
45
2. The timing of an aerial photo survey depends on the weather and seasonal development of the vegetation within the investigation area, thus, the objective of the survey determines the time of the survey. Processing and interpretation: 1. For most thematic surveys, at least two groundchecks are necessary: one for preparing the interpretation key, the other one to confirm the results of the interpretation. The ground control points for the aerial photographs and in the terrain must be recorded with comprehensive documentation; the more detail the better. 2. For interpretation, aerial photographs have to be viewed stereoscopically. For most applications, misinterpretation can be expected if the photographs are viewed only two-dimensionally. 3. High-precision aerial photo scanners should be used. 4. A rectification program for elimination of geometric image shifts must be used. 5. If results from previous investigations or other methods exist, they should be incorporated into the interpretation. An interdisciplinary approach improves the recognition and understanding of details in aerial photographs. 6. Aerial photo interpretation is limited by the specifications of the film used in the aerial photo survey, by vegetation cover, and poor weather conditions.
46
3.1 Aerial Photography
3.1.8 Personnel, Equipment, Time Needed Personnel
taking aerial photographs film development search for archived aerial photographs scanning of the aerial photographs evaluation of development of a landfill, industrial or mining site evaluation of the available data about environmental conditions and geology of a landfill, industrial or mining site mapping of seepage water at the edges of landfills evaluation of springs and soil moisture anomalies investigation of natural and artificial drainage systems
special survey aircraft with 2-3 technicians flight management system, or specialists aerial metric camera, aerial film 1 technician 1 roller transport processor 1 archivist 1 technician 1 specialist
1 specialist
1 specialist 1 specialist 1 specialist
fracture analysis
1 specialist
evaluation of destabilization of landfill or mining waste heap slopes
1 specialist
mapping of vegetation vitality
1 specialist
assessment of surface-water conditions mapping of land-use patterns and biotopes inventory of other suspected contaminated sites in the area around a landfill, industrial or mining site
Equipment
1 specialist 1 specialist 1 specialist
search for a new waste disposal site
1 specialist
ground check digitalization of one thematic map
1 specialist 1 technician
Time needed [d] 1 1 3 - 14
1 photogrammetric scanner 1 stereoscope 1 PC 1 stereoscope 1 PC 1 stereoscope 1 PC 1 stereoscope 1 PC 1 stereoscope 1 PC 1 stereoscope 1 PC 1 stereoscope 1 PC 1 stereoscope 1 PC 1 stereoscope 1 PC 1 stereoscope 1 PC 1 stereoscope 1 PC 1 stereoscope 1 PC 1 4WD vehicle 1 PC
0.5 2-5
5 - 15
0.5 0.5 - 5 1-2 5 - 15 1-5 5 - 15 1-2 5 - 15 3 - 15 3 - 15 1-5 0.5 - 5
Environmental Geology, 3 Remote Sensing
47
3.1.9 Examples Development of a waste disposal site Two aerial photographs of the Schöneicher Plan waste disposal site near Berlin, Germany, are shown in Figures 3.1-7 and 3.1-8a. They were taken in 1992 and 1967, respectively. An interpretation of the 1967 photograph by KRENZ (1991) is given in Figure 3.1-8b. These aerial photographs allow a chronological investigation of the disposal site. The areas where waste was dumped and the technology used can be identified in good quality aerial photographs, and the risks of these waste deposits can be evaluated.
Fig. 3.1-7: Panchromatic aerial photograph (photomosaic) taken on May 15, 1992, showing the areal extent of the Schöneicher Plan waste disposal site (courtesy: Luftbildsammelstelle der Landesvermessung und Geobasisinformation Brandenburg)
48
3.1 Aerial Photography
Fig. 3.1-8: (a) Aerial photograph taken on June 23, 1967; (b) Interpretation by KRENZ (1991) on a topographic base map from 1982 (source: Bundesarchiv (Federal Archives of Germany)
Environmental Geology, 3 Remote Sensing
49
Evaluation of the previous environmental and geological situation at the location of an operating landfill An aerial photograph from 1991 (Fig. 3.1-9) depicts an abandoned waste disposal site southeast of Ludwigslust in Mecklenburg-Vorpommern, Germany. The stereo-pair from 1953 (Fig 3.1-10) shows that the site was a gravel pit that has been filled with waste. Leachate from the waste site has direct access to the uppermost aquifer, which is exposed at the sides of the pit without the benefit of any natural filtering through soil. This would have occurred if the waste site had been placed on the natural surface instead of in a man-made depression. The depth of the pit (and thickness of the waste), the areal extent of the old activities and the quantity of waste were determined by interpretation of the 1953 stereo-pair. If the archived photographs had not been available, the subsurface characteristics of the waste site could have been determined using shallow geophysical techniques, such as dc resistivity and shallow seismic methods, but at much greater expense and time.
50
3.1 Aerial Photography
Fig. 3.1-9: Aerial photograph (stereo-pair) taken on July 7, 1991, showing a disused waste disposal site (arrow) southeast of Ludwigslust in Mecklenburg-Vorpommern, Germany. The presently covered and partially overgrown waste site slightly rises above the surrounding terrain (printed with courtesy of the Amt für Geoinformationswesen der Bundeswehr in Euskirchen, Germany). Luftbild © AGeoBw, 1991 - Lizenz B - 7A004
Environmental Geology, 3 Remote Sensing
51
Fig 3.1-10: Archive aerial photograph (stereo-pair) taken on May 28, 1953, showing a deep gravel pit (arrow) at the location of a now abandoned waste disposal site southeast of Ludwigslust (Fig. 3.1-9). Direct dumping of waste into the pit provides direct pathways for contaminated leachate into the uppermost aquifers (source: uve GmbH Berlin)
52
3.1 Aerial Photography
Seepage of water at the edge of a waste disposal site Anomalous plant growth, discoloration of the soil, and visible wet areas at the edge of a landfill indicate seepage of water from the landfill and potential groundwater contamination at the site (Fig. 3.1-11). In general, water seeping from of a landfill is likely to be highly contaminated. Once leachate from a landfill is identified, further action is needed to characterize the leachate and control or restrict its movement.
Fig. 3.1-11: CIR aerial photograph taken on August 2, 1990, showing signs of water leachate (within the dashed rectangle) near the edge of an abandoned disposal site which was transformed into a park in Berlin (aerial photograph: Eurosense GmbH, printed courtesy of the Senat Department for Urban Development, Berlin)
Environmental Geology, 3 Remote Sensing
53
Areas of high and low permeability to water Plant growth depends, among other factors, on the permeability of the soil and the underlying beds. The lighter colored areas in Figure 3.1-12 correlate with areas of poor plant growth due to the sandy soil and the permeable beds there. Precipitation and leachate from the neighboring landfill can migrate through this permeable material directly into the aquifer.
Fig. 3.1-12: CIR aerial photograph taken on July 3, 1993, of an area west of the Schöneiche waste disposal site. Extensive sandy soil substrates responsible for impaired plant growth can be recognized by the lighter colors, (Aerial photograph: WIB GmbH for BGR).
54
3.1 Aerial Photography
Soil moisture anomalies, natural and artificial drainage systems It is often a coincidence if archived historical aerial photographs can contribute to the clarification of site-specific problems. Seasonal conditions and the weather at the time the aerial photographs were taken determine their usability. Weather conditions during a reconnaissance flight in the Mittenwalde area south of Berlin on April 10, 1945 have been optimal (Fig. 3.1-13). New vegetation growth patterns delineated substrate and moisture conditions in the area at the time the data was acquired. Currently, the area of faint stripes recognizable in Figure 3.1-13 is covered by the northern part of the Schöneiche landfill. The stripes may indicate long-term agricultural use (old field patterns) or the existence of abandoned drainage systems (see SCHNEIDER, 1974). Although no old drainage systems were documented in this area, it cannot be excluded that the faint linear features (arrow) indicate an old artificial drainage system. A ground check of such features is recommended if they are observed within the area of a landfill. An abandoned drainage system may still function and may transport waste leachate away from the landfill. In this particular case, damaged plants and slightly discolored water were observed in CIR aerial photographs of the areas where the feature joins the Gallun Canal (Fig. 3.1-16). The area marked by the blue circle in Figure 3.1-14 attracted attention during a ground check because of an abnormal phenotype of a large tree. At this time every other tree of the same species around the site had brown, ripe fruits with only small water content. But, the fruit on this tree were still green and juicy. The tree is at the top of the slope at the edge of the landfill. Closer inspections revealed water seeping from the slope even during the dry season. Interviews with the residents of a nearby village, together with the interpretation of the archival aerial photographs taken at the end of World War II, confirmed the existence of former irrigation ditches. Obviously the ditches are, at least to some extent, still functional beneath the covering waste.
Environmental Geology, 3 Remote Sensing
55
Fig. 3.1-13: Wartime aerial photograph taken on April 10, 1945, showing cultivated land now covered by the northern part of the Schöneiche landfill (within dashed line). The high contrast reveals differences in moisture content and lithology. Faint stripes (arrow) suggest the presence of an old drainage system or long-term agricultural use (old field patterns). (Courtesy of Luftbilddatenbank, Ing.-Büro Dr. Carls, Estenfeld)
56
3.1 Aerial Photography
Fig. 3.1-14: Archive aerial photograph taken on February 1, 1945, showing a rice field near the city of Chiang Mai in northern Thailand. Irrigation ditches can be observed in the photograph within the rice field, which is now a waste disposal site. The ditches act as drainage system leading contaminants to the edge of the present-day landfill. Dashed red line: boundary of the waste disposal site. Arrow: direction of flow in the rice field’s irrigation ditches. Blue circle: wet area at the edge of the waste site with assumed leachate. Archives research and acquisition by Luftbilddatenbank, Ing.-Büro Dr. Carls, Estenfeld, for BGR
Destabilization of mining spoil and tailings heap slope Under certain geological conditions, the potential risk is not limited for soil and groundwater contamination. The stability of waste disposal sites and mining spoil and tailings heaps may be threatened as well. The stereo-pair of aerial photographs in Figure 3.1-15 shows part of a tailings heap south of Magdeburg, Germany. This heap (the white area on the left) contains waste from potash production. A sinkhole caused by collapse in the potash mine appeared near the heap in 1975 (LÖFFLER, 1962 and BRÜCKNER et al., 1983). Stable conditions apparently prevail on the opposite side of the sinkhole from the heap (no concentric collapse fractures). However, the terrain surface is gradually collapsing along concentric fractures in the direction of the heap. Aerial photographs taken over a period of time can be interpreted to determine the stability of the area and predict additional collapse (KÜHN et al., 1997 and 1999).
Environmental Geology, 3 Remote Sensing
57
Fig. 3.1-15: Panchromatic aerial photographs (stereo-pair) can be interpreted to predict whether a waste heap will become unstable. Circular features at the edge of the collapse area (diameter about 270 m) indicate ongoing collapse in the direction of the heap (Photograph taken May 8, 1994, by Berliner Spezialflug, Luftbild GmbH for BGR).
58
3.1 Aerial Photography
Vitality of vegetation A CIR photograph and map depicting the state of health of trees in the vicinity of the Schöneicher Plan and Schöneiche landfills are shown in Figures 3.1-16 and 3.1-17. The landfills are 0.5 m below and 1.5 m above the groundwater level, respectively. The uppermost, unconfined aquifer consists of sand and is in hydraulic contact with a number of streams. Runoff from both sites mainly passes through numerous ditches and finally flowing into the Notte Canal, to the north of the sites. The lack of an impermeable layer, e.g., clay, acting as a seal at the base of the landfills, allows the leachate to enter the aquifer and, thus, the surface drainage system. The CIR image shows the area north of the Schöneiche landfill with several rows of trees growing along both the open and silt-filled ditches. Most of the trees are willows, black poplars, and alders. The map (Fig. 3.1-17) was compiled using only CIR aerial photographs from July 2, 1993, and ground check information. This map supports the earlier assumption that contamination spreads northward from the landfill. It can also be seen that the black poplars along the Gallun Canal northeast of the landfill are considerably stressed. This indicates that seepage water from the landfill infiltrates the ground via an old drainage system. However, an impact on the health of the black poplars by the neighboring waste incineration plant cannot be excluded.
Environmental Geology, 3 Remote Sensing
59
Fig. 3.1-16: CIR aerial photograph taken on July 2, 1993, showing an area north of the Schöneiche landfill. The road along the northern edge of the landfill is just visible at the right bottom of the photo (photo taken by WIB GmbH Berlin for BGR).
60
3.1 Aerial Photography
Fig. 3.1-17: Map of tree vitality near the Schöneicher Plan and Schöneiche landfills; mapped at a scale of 1 : 5000 (simplified)
Environmental Geology, 3 Remote Sensing
61
Assessment of surface water conditions In many cases, pollution of surface water is indicated by changes in the color of the water in the CIR photographs. The blue of the water body in the CIR aerial photograph of Figure 3.1-18 indicates intensive turbidity. This turbidity was determined by a ground check to be caused by illegal discharge of liquid manure from a nearby farm.
Fig. 3.1-18: CIR aerial photograph of water-filled clay pits of a former brickyard north of Ketzin, Brandenburg, Germany. The blue color indicates the discharge of liquid manure (water bodies are normally dark or black in CIR photographs). Photograph taken July 28, 1990, by Berliner Spezialflug, Luftbild GmbH; printed with courtesy of FUGRO Consult GmbH, Berlin.
62
3.1 Aerial Photography
Fractures Disturbed ground and fractures are often easy to detect in aerial photographs. The surface expression of such features is marked by changes in topography, drainage patterns, vegetation growth patterns, and rock and soil color differences. A laser image (Chapter 3.3) and CIR aerial photograph of the same area are shown in Figure 3.1-19. The CIR aerial photograph was taken on April 21, 2000, and shows a field of young cereal (Fig. 3.1-19B). Due to the spectral sensitivity of the film, the red color indicates healthy vegetation. In this photograph, the intensive red indicates advanced growth of the young plants. The vitality of the plants is due to the fertile soil, composed of thick loess, accumulated in open fractures. The red lines in the aerial photograph mark the individual fractures. Narrow linear depressions visible in the laser image (Fig. 3.1-19A) suggest that the ground is already subject to displacement. The two images in Figure 3.1-19 corroborate the existence of a fracture zone, which was also verified in the field. Since this area is still used for farming, potential hazards to people, road traffic, and farm machinery have to be taken into consideration.
Fig. 3.1-19: “Laser Image” (A, shaded relief map – sun elevation: 45o, azimuth: 45o) and B Color-infrared (CIR) aerial photograph of the same part of a fracture zone north of the lake “Suesser See” near Eisleben, Germany. Faint linear features in the laser image and linear to curvilinear features in the CIR aerial photograph (yellow arrows) are surface expressions of fractures.
Environmental Geology, 3 Remote Sensing
63
Mapping of land use pattern and biotopes In many countries, land use maps are used to assist planning and decisionmaking at the regional scales. The CIR photograph in Figure 3.1-20 depicts an industrial plant and its waste disposal site. The landfill was established for temporary storage of ferruginous dust and mud. Two mud ponds and a thin vegetation cover are present in the southern part of the landfill. No vegetation can develop where the dust was deposited. The land use interpretation in Figure 3.1-21 documents the problematic location of the waste disposal site with respect to the nearby settlement. Residential buildings are less than 100 m away from this source of dust emission (red area). Dust transported by southerly and easterly winds may have negative affects on the health of the residents and the agriculture fields in the vicinity of the waste site. Proper information and planning prior to the establishment of a waste disposal site helps to avoid such problems.
Fig. 3.1-20: CIR aerial photograph of an industrial plant (lower right corner) and its waste disposal site (in the center) near the city of Esch-sur-Alzette in Luxembourg. The photo was taken on May 27, 1999, at a scale of 1 : 15 000 by Hansa Luftbild (German Air Survey), Münster
64
3.1 Aerial Photography
Fig. 3.1-21: Interpretation of aerial photographs with regard to land use as a basis for regional planning, Hansa Luftbild (German Air Survey), Münster
Land use mapping Land use was mapped during an investigation of the abandoned Nong Harn landfill near Chiang Mai (northern Thailand) using the classification system of WACHARAKITTI (1982) adapted to the objectives of the site investigation and using 1 : 7500 CIR aerial photographs (Fig. 3.1-22). The classification system distinguishes six hierarchic levels of land categories. The first hierarchical level is labeled by letters, e.g., “a” for agricultural land, ”u” for urban land, and “f” for forest land; the sublevels are labeled by numbers, separated by periods. If in one level an attribute cannot be assigned to an area, a blank is used instead of a number. The more attributes that can be recognized, the better the land use of the area can be specified.
Environmental Geology, 3 Remote Sensing
65
Fig. 3.1-22: Land use map of the Nong Harn area containing the codes of all objects recognized in the interpretation of aerial photographs
66
3.1 Aerial Photography
Inventory of sites suspected to be hazardous An inventory of sites suspected to be hazardous in the surroundings of the abandoned Mae Hia landfill, Chiang Mai, Thailand was performed using historical aerial photography. The goal was to assess the different possible sources of contamination. Due to the poor records available, the interpretation scheme was limited to two classes. Sites and facilities with likely existence of hazardous substances have been assigned to class 1. Less dangerous sites were mapped as class 2. Table 3.1-4 and Figure 3.1-23 show the results of the inventory. Table 3.1-4: Overview of the sites suspected to be hazardous in the Mae Hia investigation area Class 1
Class 2
Abandoned industrial sites motor vehicle garage gas/petrol station workshop in which wood is impregnated or is worked other Waste deposits
2 2 1
landfills small sites with buried and/or heaped waste litter
2
19
9 89
Fig. 3.1-23 (next page): Sites in the Mae Hia area of northern Thailand suspected to be contaminated on the basis of interpretation of aerial photographs
Environmental Geology, 3 Remote Sensing
67
68
3.1 Aerial Photography
References and further reading ALBERTZ, J. (2001): Einführung in die Fernerkundung, Grundlagen der Interpretation von Luft- und Satellitenbildern. Wissenschaftliche Buchgesellschaft, Darmstadt. AVERY, T.E., Berlin GL (1985): Interpretation of aerial photographs. Burgess Publ. Comp., Minneapolis. BALTASAVIAS, P. & KAESER, C. (1999): Quality Evaluation of the DSW 200, DSW 300, SCAI and Ortho Vision Photogrammetric Scanners. Unpublished Report. BARRETT, E.C. & CURTIS, L. F. (Eds.) (1974): Environment remote-sensing: applications and achievements. Edward Arnold Ltd., London. BÖKER, F. & KÜHN, F. (1992): Zur Verwendung von Luftbildern bei der geologischen Kartierung. Zeitschrift für angewandte Geologie, 38, 2, 80-85. BORMANN, P. (1981a): Passive Fernerkundungssensoren im optischen Bereich des Spektrums. Vermessungstechnik, 29, 2, 45-48. BORMANN, P. (1981b): Was ist unter dem Auflösungsvermögen eines Fernerkundungssensors zu verstehen. Vermessungstechnik, 29, 10, 331-335. BORRIES, H.-W. (1992): Altlastenerfassung und -erstbewertung durch multitemporale Karten- und Luftbildauswertung. Vogel, Würzburg. BRÜCKNER, G., KNITSCHKE, G., SPILKER, M., PELZEL, J., SCHWANDT, A. (1983): Probleme und Erfahrungen bei der Beherrschung von Karsterscheinungen in der Umgebung stillgelegter Bergwerke des Zechsteins in der DDR. Neue Bergbautechnik, 13, 8, 417-422. BRUNEAU, M. (1980): Mapping Landscape Dynamics in the Highlands and Lowlands of Northern Thailand. In: IVES, J. D., SABHASRI, S. & VORAURAI, P.: Conservation and Development in Northern Thailand, The United Nations University, Tokyo. Center for Remote Sensing and Geographic Information Science (1999): Historical Aerial Photos and Maps for Documenting Changes over a Site. Michigan State University, www.crs.msu.edu. CHUCHIP, K. (1997): Satellite Date Analysis and Surface Modeling for Land Use and Land Cover Classification in Thailand. Berliner Geographische Studien, 46, Berlin. CICIARELLI, J. A. (1991): A Practical Guide to Aerial Photography. Van Nostrand, Würzburg. COLWELL, R. N. (ed.) (1983): Manual of Remote Sensing, 1, Theory, Instruments and Techniques, 2, Interpretation and Applications. American Society of Photogrammetry, Falls Church. DECH, S. W., GLASER, R., KÜHN, F. & CARLS, H.-G. (1991): Ökologische Probleme durch Rüstungsaltlasten in der Colbitz-Letzlinger Heide. DLR-Nachrichten, 64. DE MILDE, R. & SAYN-WITTGENSTEIN, L. (1973): An Experiment in the Identification of Tropical Tree Species on Aerial Photographs. Proc. IUFRO Symp. S. 6.05, Freiburg 1973, 21-38.
Environmental Geology, 3 Remote Sensing
69
DIN 18740-1 (1999): Qualitätsanforderungen an photogrammetrische Produkte, Teil 1: Bildflug und analoges Luftbild, Deutsches Institut für Normung e.V. DODT, J., BORRIES, H. W., ECHTERHOFF-FRIEBE, M. & REIMERS, M. (1987): Zur Verwendung von Luftbildern und Karten bei der Ermittlung von Altlasten. RuhrUniversität Bochum. DOUGLAS, S. (1973): Terrain Analysis – A Guide to Site Selection Using Aerial Photographic Interpretation. Community Development Series, Dowden, Hutchinson & Ross Inc., Stroudsburg. EDWARDS, G. J. (Ed.) (1988): Color Aerial Photography in the Plant Sciences and Related Fields, A Compendium 1967 - 1983, Falls Church. EHRENBERG, M. (1991): Beprobungslose Altlastenerkundung. wlb. Wasser, Luft und Boden, 7-8, 56-59. ERB, W. (1989): Leitfaden der Spektroradiometrie. Springer, Berlin. GLASER, R. & CARLS, H.-G. (1990): Kriegsluftbilder 1940 - 1945: Ein Hilfsmittel bei der Verdachtsflächenermittlung von Kriegsaltlasten und in der Umweltplanung. Laufener Seminarbeitr., 90.1, 65-73, Laufen/Salzach. GOETZ, A. F., BARRET, N. R. & ROWAN, L. C. (1983): Remote Sensing for Exploration: An Overview. Econom. Geol., 78/4, 573-592. GUPTA, R. P. (1991): Remote Sensing Geology. Springer, Berlin. HAAS, R. (1992): Konzepte zur Untersuchung von Rüstungsaltlasten. Erich Schmidt, Berlin. HEGG, K. M. (1966): A photo identification guide for the land and forest types of interior Alaska. U.S. For. Serv. North. For. Exp. Station, Research Paper NOR-3, 55. HILDEBRANDT, G. (1996): Fernerkundung und Luftbildmessung: für Forstwirtschaft, Vegetationskartierung, und Landschaftsökologie. Herbert Wichmann, Heidelberg. HOLZFÖRSTER, B. & TIEDEMANN, M. (1991): Die Bedeutung der Luftbildauswertung für die Erfassung von Rüstungsaltlasten am Beispiel Niedersachsen. In: THOMÉ-KOZMIENSKY (Ed.): Untersuchung von Rüstungsaltlasten. EF-Verlag für Energie und Umwelttechnik, Berlin. HUBER, E. & VOLK, P. (1986): Deponie- und Altlastenerkundung mit Hilfe von Fernerkundungsmethoden. Wasser und Boden, 10, 509-515. KENNEWEG, H. (1994): Forest condition and forest damages – Contribution of remote sensing. Geo-Journal, 32, 47-53. KRENZ, O. (1991): Luftbildinterpretation der Deponiestandorte Schöneiche und Schöneicher Plan - Teilbericht zum BMFT-Projekt „Abfallwirtschaftliche Rekonstruktion von Altdeponien am Beispiel des Deponiestandortes Schöneiche“. Unpublished report. KRONBERG, P. (1984): Photogeologie, eine Einführung in die Grundlagen und Methoden der geologischen Auswertung von Luftbildern. Enke, Stuttgart. KRONBERG, P. (1985): Fernerkundung der Erde. Enke, Stuttgart.
70
3.1 Aerial Photography
KÜHN, F. & OLEIKIEWITZ, P. (1983): Die Nutzung der Multispektraltechnik zur Früherkennung von senkungs- und erdfallgefährdeten Gebieten. Z. angew. Geol., 29, 6, 71-74. KÜHN, F. & HÖRIG, B. (1995): Geofernerkundung, Handbuch zur Erkundung des Untergrundes von Deponien und Altlasten, 1, Springer, Heidelberg. KÜHN, F., HÖRIG, B. & BUDZIAK, D. (2004): Detecting unstable ground by multisensor remote sensing. Photogrammetrie – Fernerkundung – Geoinformation (PFG), 2, 101-109. KUEHN, F., KING, T., HOERIG, B. & PETERS, D. (2000): Remote Sensing for Site Characterization. Methods in Enviromental Geology, Springer, Berlin. KÜHN, F., TREMBICH, G. & HÖRIG, B. (1997): Multisensor Remote Sensing to Evaluate Hazards Caused by Mining. In: Proceedings of the Twelfth International Conference on Applied Geologic Remote Sensing, 17-19 November, Denver, Colorado, ERIM, I, 425-432. KÜHN, F., TREMBICH, G. & HÖRIG, B. (1999): Satellite and Airborne Remote Sensing to Detect Hazards Caused by Underground Mining. In: Proceedings of the Thirteenth International Conference on Applied Geologic Remote Sensing, 1-3 March, Vancouver, British Columbia, Canada, ERIM, II, 57-64. LINTZ, J. & SIMONETT, D. S. (Eds.) (1976): Remote Sensing of Environment. AddisonWesley, London. LÖFFLER, J. (1962): Die Kali- und Steinsalzlagerstätten des Zechsteins in der Deutschen Demokratischen Republik. Freiberger Forschungsheft, Teil III, Sachsen-Anhalt, C97/III, Akademie Verlag, Berlin. MILLER, V. & MILLER, C. F. (1961): Photogeology. McGraw-Hill, New York. MURTHA, P. A. (1972): A Guide to Air Photo Interpretation of Forest Damage in Canada. Can. Forestry Service Publ., 1292, 63. MYERS, B. J. (1978): Separation of Eucalyptus species on the basis of crown colour on large scale colour aerial photographs. Austr. For. Res., 8, 139-151. ONGSOMWANG, S. (1994): Forest Inventory, Remote Sensing and GIS (Geographic Information System) for Forest Management in Thailand. Berliner Geographische Studien, Berlin. PAINE, D. P. (1981): Aerial Photography and Image Interpretation for Resource Managment. Wiley & Sons, New York. PETERS, D. C., HAUFF, P. L. & LIVO, K. E. (1995): Remote sensing for mine waste discrimination and characterization. In: CURRAN, P. J. & ROBERTSON, Y. C., comps., RSS 95: Remote Sensing in Action, Southampton, UK, Sept. 12 - 15, 1995, The Remote Sensing Society, 866-877. SABINS, F. F. (1996): Remote sensing - Principles and interpretation. 3rd edition, Freeman, San Francisco.
Environmental Geology, 3 Remote Sensing
71
SAYN-WITTGENSTEIN, L. (1960): Recognition of tree species on air photographs by crown characteristics. Com. Dept. For. Techn. Note 95. SCHMIDT, D. (2000): Vitality of Trees. In KUEHN, F., KING, T., HOERIG, B. & PETERS, D. (Eds.): Remote Sensing for Site Characterization, Springer Verlag, Berlin, 91-93. SCHNEIDER, S. (1974): Luftbild und Luftbildinterpretation. de Gruyter, Berlin. SCHNEIDER, S. (1984): Angewandte Fernerkundung: Methoden und Beispiele. Vincentz, Hannover. SCHWEBEL, R. (2001): Qualitätssicherung für Bildflug und analoges Luftbild durch neue DIN-Norm, Photogrammetrie, Fernerkundung, Geoinformation, 1, 39-44. SMITH, J. T. & ABRAHAM, A. (Eds.) (1968): Manual of Color Aerial Photography. American Society of Photogrammetry, Falls Church. STRATHMANN, F.-W. (1993): Taschenbuch zur Fernerkundung. Wichmann, Karlsruhe. TANHAN, S. (1989): Application of Aerial Photography for Land Right Cultivation. In: Remote Sensing and Mapping for Forest Management: Report of Project Course B.UNDP, FAO, Bangkok, 261-264. TEEUW, R. M. (2007): Mapping hazardous Terrain using remote sensing. GSL Special Publications, London. THORLEY, G. ed. (1975): Forest Lands: Inventory and Assessment. In: Manual of Remote Sensing 1975, II, American Society of Photogrammetry. TIWARI, K. P. (1975): Tree species identification on large scale aerial photographs at new forest. Indian forest, 10, 791-807. VDI-Guideline 3793 (1990): On-site Determination of Vegetational Injuries Method of Aerial Photography with Color Infrared Film. Verein Deutscher Ingenieure, VDIHandbuch Reinhaltung der Luft, 1, Berlin. WACHARAKITTI, S. (1982): Land Use Classification System or Land Use Design. Faculty of Forestry, Kasetsart University, Bangkok (Version Thai). WEBER, H. (ed.) (1993): Altlasten: Erkennen, Bewerten, Sanieren. Springer, Berlin.
Environmental Geology, 3 Remote Sensing
73
3.2 Photogrammetry DIETMAR SCHMIDT, PETER HERMS, ANKE STEINBACH & FRIEDRICH KÜHN
3.2.1 Principle of the Methods Photogrammetric techniques provide reliable measurements of geometric characteristics of Earth surface features from photographic images taken from remote sensing platforms. These remote measurements not only include the size, shape, and position of objects, but also their color or tone, texture, and spatial patterns and associations. Hence, such observations can be interpreted and analyzed with regard to their geoscientific information (Chapter 3.1) and with regard to their geometric structures (Chapter 3.2). Photogrammetry techniques are used to prepare maps and digital elevation models (DEM) from remote-sensing data. Other tasks are the preparation of orthophotographs, photomosaics, terrain cross-sections, and vector data (e.g., the height of a terrain slope, length of a stream, volume of a waste heap or landfill). Most photogrammetric products derived from aerial photographs have been used for decades. These comprehensive experiences have developed to standards and rules assuring the quality and consistency of the mapping products (SCHWEBEL, 2001; STUTTARD & DOWMAN, 1998). There are alternative approaches to traditional photogrammetry. For obtaining topography data airborne laser scanning has proven to be a successful method (Chapter 3.3). Optical-electronic cameras (Chapter 3.3) have also been developed for photogrammetric purposes. These sensor systems provide geometric resolution down to the centimeter range. Digital cameras that meet the requirements for photogrammetry processing are in an experimental stage but are expected to widely replace conventional aerial cameras in the near future (Chapter 3.3). The major objective of photogrammetry is to relate the pixel coordinates measured by the sensor as exactly as possible to the geographic coordinates (longitude, latitude, height) of terrain points. This process involves the removal of distortions caused by the data acquisition system (aerial camera or scanner), perspective, and motion of the aircraft or the space-borne platform. This processing, called rectification of the image, requires the specifications of the camera and precise coordinates from ground control points (GCP). Two types of digital processing are used in photogrammetry: x Geometric correction: This interpolation method uses northing and easting, (latitude and longitude respectively) of known terrain points and their corresponding image coordinates in the aerial photographs to obtain a "true-to-position" representation without consideration of location differences in the third dimension (topography).
74
3.2 Photogrammetry
x Orthorectification: A mathematical process in which transformation parameters, including the sensor specifications (interior orientation) and position during the acquisition of the photograph (exterior orientation), are used to obtain the geometric relationships in the aerial photographs and elevation data and thus remove all geometric distortions caused by the data acquisition process. In both cases, ground checks are essential to determine the geometric relationship between point measurements in the photographs and the true location on the ground. Ground control points should be well defined in the photograph and easy to find and accessible on the ground. Points marked by reflective foil, fabric, plastic or paint, called ground targets, may be applied during the data acquisition to provide such control points if landscape characteristics (e.g., large uniform agricultural areas, steppe or desert) are not appropriate. The coordinates of the ground control points have to be surveyed on the ground if they cannot be accurately extracted from topographic maps. These field surveys are usually quickly and inexpensively accomplished using GPS technology (Chapter 2.5). Orthophotographs, photomosaics, thematic maps, terrain cross-sections, and vector datasets derived from photogrammetry are widely used for site investigations in conjunction with other geological, hydrological, geophysical, and remote-sensing data. Topographic measurements acquired at different times can be used to observe changes of landfills, waste heaps and slopes suspected to be instable, as well as for planning and decision-making by governmental agencies and in the business world.
3.2.2 Applications x Preparation of orthophotographs and photomosaics, x development of topographic maps, x derivation of digital elevation models (DEM), x derivation of terrain cross-sections and vector data (e.g., determination of the height of a terrain slope, length of a stream, volume of a waste heap or landfill).
3.2.3 Fundamentals The most common task for photogrammetry is the preparation of maps and comparable products, e.g., orthophotographs. While all points on a map are usually depicted at their true relative horizontal positions (parallel projection), an aerial photograph is distorted in a regular geometric manner. Because of
Environmental Geology, 3 Remote Sensing
75
these inherent geometric distortions aerial photographs cannot be directly used for mapping purposes. However, the geometric distortions can be removed or at least minimized with specific rectification procedures. The geometric distortions result from the data acquisition process as the Earth surface is being viewed from a single point (central projection), and changes in terrain elevations, as well as, the instability of the aircraft causing dislocations of image pixel coordinates. The ladder, rather irregular distortions are minimized by compensation systems mounted on the sensor platform, e.g., Image Motion Compensation (IMC) and Forward Motion Compensation (FMC), and neglected in routine investigations. Geometric distortions of vertical aerial photographs Distortions resulting from a non-vertical position of the optical axis in vertical aerial photography (rarely greater than 5 gon (4.5°) and usually much smaller) is neglected in most cases (LILLESAND & KIEFER, 1994; ALBERTZ, 2001). Because of the nature of central projection, any variation in terrain elevation will result in scale variations and displaced positions of ground objects in a vertical aerial photograph. Corresponding to Equation (3.1.1) the scale of an aerial photograph is directly proportional to the focal length cf of the camera lens and inversely proportional to the height hg of the camera above ground. Therefore, aerial photographs taken over terrain of varying elevation will have a continuous range of scales associated with changes of hg due to variations in terrain elevations. The location displacement of a ground object in a vertical aerial photograph due to variations of terrain and/or object height (topography) and central projection is called relief displacement (SABINS, 1996; GUPTA, 2003). The amount of relief displacement d in an aerial photograph is directly proportional to the difference 'h in terrain elevation and/or the object height with respect to the reference plane (map plane), directly proportional to the radial distance r from the principal point, i.e. the optical center of the photograph, and inversely proportional to the height hg of the camera above the terrain: d
'h r . hg
(3.2.1)
The maximum displacement is at the corners of the photograph. Figure 3.2–1 demonstrates how elevation differences are reflected in a horizontal displacement. Terrain depressions such as valleys and pits are shifted towards the principal point N of the image. In contrary, elevated areas such as hills, mountains, buildings, towers or walls are displaced away from the principal point.
76
3.2 Photogrammetry
Fig. 3.2-1: Radial distortion caused by relief, SCHNEIDER (1974)
Determination of object heights from radial distortion or shadow length Equation (3.2.1) can be used to determine the height of an object after solving for 'h. If the base and top of an object (building, tower, steeple, tree) are visible in the image, the object height 'h can be determined from the radial displacement 'r' (Fig. 3.2–2) using Equation (3.2.2) (ALBERTZ, 2001).
Fig. 3.2-2: Radial distortion due to differences in the height of the objects, ALBERTZ (2001)
Environmental Geology, 3 Remote Sensing
77
Fig. 3.2-3: Determination of the heights of objects from shadow length. Left: in the case of a known sun elevation angle D. Right: if the object height 'h1 is known, ALBERTZ (2001)
'h
'r ' h g r'
(3.2.2)
The height of the object can be estimated from the length l' of its shadow if the angle D with respect to the position of the sun (sun elevation angle) is known (Fig. 3.2-3, left) (SCHNEIDER, 1974; ALBERTZ, 2001). Sn is the scale constant (Equation 3.1.1). 'h
l ' S n tan D
(3.2.3)
If the height 'h1 of one object is known, the height 'h2 of other objects (Fig. 3.2-3, right) can be determined using Equation (3.2.4) (ALBERTZ, 2001): 'h2
l 2' 'h1 . l1'
(3.2.4)
Simple rectification of vertical aerial photographs A simple rectification procedure can be applied in case of flat terrain. As a rule of thumb, a terrain can be regarded as flat if the elevation differences are less than the scale constant Sn (Equation 3.1.1) divided by 500. Table 3.2-1 gives the elevation differences for which terrain can be regarded as being flat, and the resulting position errors as a function of map scale.
78
3.2 Photogrammetry
Table 3.2-1: Permissible elevation differences for which a terrain can be regarded being flat and resulting position errors as a function of map scale (LÜSCHER, 1944; SCHNEIDER, 1974) Map scale
Permissible elevation difference in m
Resulting position error in m
1 : 1000 1 : 2000 1 : 5000 1 : 10 000 1 : 25 000 1 : 50 000
2 4 10 20 50 100
1 2 5 10 25 50
Graphical, optical-photographic, and digital image processing methods are available for rectification of aerial photographs in areas with flat terrain (ALBERTZ, 2001). The perspective relationship between the plane of the image and the ground surface can be unambiguously determined from four ground control points with known coordinates (e.g., from GPS measurements) that can be accurately located in the photograph. In digital image processing the relationship between the point locations on the image plane and on the ground surface can be expressed by two equations. The coefficients in the equations are determined from the ground control points and solved respectively. Stereometric processing of vertical aerial photographs Aerial photographs successively taken along a flight line usually overlap by approximately 60 %. The area of overlap allows for stereoscopic viewing, i.e., viewing the two images simultaneously using stereoscopic instruments. Viewed with the proper spacing of the images, the area is presented in three dimensional perspective. The photographs must have about the same scale, i.e. they must be taken from a similar sensor flight elevation. Satellites, e.g., IKONOS and QuickBird-2 have the capability to tilt the line-of-sight in either the along-track or the across-track direction to generate both along-track and across-track stereo pairs of photographs (GUPTA, 2003). Stereoscopic viewing and measurement is based on the “principle of parallax”, which is fundamental to photogrammetric processing. Parallax is the apparent displacement of the position of stationary objects caused by a shift in the position of the viewer, e.g., apparent location difference of a target in two consecutive, overlapping aerial photographs. In vertical aerial photographs, parallax displacements only occur parallel to the flight line. Distortions due to aircraft instabilities affect the compatibility of photographs for stereometric processing. Figure 3.2-4 illustrates the concept of parallax and the geometric relations in stereoscopic viewing for overlapping vertical aerial photographs.
Environmental Geology, 3 Remote Sensing
79
Fig. 3.2–4: Concept of Parallax: stereoscopic viewing of overlapping aerial photographs and geometric relationships, SCHNEIDER (1974)
For stereoscopic processing, the first photograph in the flight direction has to be aligned on the left side of the stereoscope and the overlapping photograph on the right side. The image centers (nadir points) N'1 and N"2 of the aerial photographs have to be determined and the nadir point in the left photograph is marked on the right one and vice versa (Fig. 3.2-4). The flight line axis can be found by connecting the nadir points N'1 and N"2. Point A on the ground is represented in the stereo pair of photographs as A' and A". The distances from nadir points N'1 and N"2 to the image points A' and A" are x' and x". The difference p = x' - x" is called the parallax of point A. For the parallax p the Equation (3.2.5) can be established: p f
b . h
(3.2.5)
80
3.2 Photogrammetry
The focal length of the camera lens is f, the height of the camera above a point on the ground is h. The horizontal distance b between the exposure points N1 and N2 is called the air base. Equation (3.2.5) can be solved with respect to h, as shown in Equation (3.2.6), and be used to estimate the elevation of the terrain: h
fb . p
(3.2.6)
The height 'h (in meters) of an object can be determined from measurements of the parallax difference 'p in the positions of the top and base of the object measured in hundredths of a millimeter: 'h
h0 'p , b ' 'p
(3.2.7)
where b' is the horizontal distance between points N'1 and (N'2) in Figure 3.2-4 measured in hundredths of millimeter range, and h0 is the flight elevation above ground in meters. Measurements of object height and elevation differences are made with stereoscopic instruments. Numerous instruments such as lens stereoscope, mirror stereoscope, and analytic data processing systems are available but differ in their technical capabilities. In stereometric processing the threedimensional coordinates with respect to a reference plane can be calculated for any terrain point and/or object. These relative coordinates are transformed using ground control points into the (absolute) coordinates of the national or international topographic reference system 1 . Photogrammetric processing is predominantly carried out in a digital environment. Automatic digital stereoscopic restitution is carried out in three steps: 1. Calculation of the spatial coordinates and parallax: An automatic recognition process for image correlation searches for and determines identical (corresponding) points in a pair of scanned stereoscopic images. Corresponding image points are identical terrain points in both images of a stereo pair. The spatial coordinates of these points and their parallax are calculated. 2. Introduction of ground control point coordinates and development of a true position model: The parallax values only reflect relative height differences. Thus, ground control point coordinates (from GPS measurements, topographic maps, etc.) are to be included in the dataset. They allow
1 The most important reference systems for worldwide applications are the UTM-system
(Universal Transverse Mercator system), the MGRS (UTMREF) for military operations and the UPS-system for polar regions.
Environmental Geology, 3 Remote Sensing
81
determination of absolute heights and are used to derive a true position elevation model. 3. Interactive modifications: The elevation model is improved interactively by editing areas with no or incorrect data. Objects which are not required (e.g., houses, trees) are eliminated. Further detail for photogrammetric processing is given, e.g., by HALLERT (1968); ALBERTZ & KREILING (1989); WOLF & WITT (2000); EGELS & KASSER (2001). Digital elevation models (DEM) can also be prepared by digital stereoscopic restitution. Figure 3.2-11 illustrates the results of the digital stereoscopic restitution of a landfill. Differential rectification of vertical aerial photographs If displacements caused by large elevation differences exceed a specified tolerance (see Table 3.2-1), differential rectification has to be applied to transform the data and correct for these distortions. This process eliminates displacement caused by topography resulting in an orthophotograph. Such rectification is more complex and more expensive than simple interpolation and requires detailed elevation data, typically an elevation model. Detailed elevation data are usually derived by digitizing contour lines from existing topographic maps. If such maps are not available or are insufficient, elevation information has to be determined by stereo photogrammetric restitution of the aerial photographs or by other methods such as airborne laser scanning. Ground control point coordinates and the camera specifications described in a calibration certificate (provided by the company that conducts the aerial photographic survey) are required in addition to elevation data.
3.2.4 Instruments Common remote sensing instruments used in photogrammetry are described in Section 3.1.4. Table 3.2-2 gives an overview of the technical specifications of photogrammetric scanners. Photogrammetric scanners are used for digitizing analog aerial photographs for digital image processing. In addition, airborne laser scanners are used as an alternative to traditional aerial photography for three-dimensional imaging of terrain.
82
3.2 Photogrammetry
Table 3.2-2: Types of photogrammetric scanners, modified after BALTSAVIAS & KAESER (1999) Name
LH Systems/DSW 500
manufacturer
LH Systems, USA
scan pixel size [ȝm] radiometric resolution [bit] (10 bit internal/ 8 bit output) scanning format x/y [mm] roll film width/length [mm/m]
ISM XL–10
4 - 20
Swissphoto Vermessung AG, Switzerland 10 - 320
10/8 or 10
10/8
265/265 35 - 241/152 manual, automatic
ZI/Imaging PhotoScan 2001 ZI Imaging, Germany 7 - 224 10/8
254/254 275/250 211 245/150 manual, automatic manual, automatic
3.2.5 Survey Practice Aerial photographic survey practice is described in Section 3.1.5 and survey practice of nonphotographic imaging systems is described in Section 3.3.5.
3.2.6 Processing and Interpretation of Data Derivation of orthophotographs, photomosaics and topographic maps Orthophotographs are determined from aerial photographs using a perspective projection where displacements from the tilt and topography have been removed. Once projected, orthophotographs have a consistent scale throughout the image and can be used as for mapping purposes. An assembly of several overlapping and/or adjacent aerial photographs, satellite or scanner images joined together to a continuous spatial representation is called a photomosaic. The preparation of orthophotographs and photomosaics starts with the scanning of aerial photographs and ground checks of control points using GPS or other reference data. Aerial photographs with scales finer than or equal to 1 : 15 000 should be used. The resolution of photographs at coarser scales is usually too low. Optimal definition of the scanning resolution has to consider that at least four pixels are needed to clearly identify an object. As a rule of thumb, this means: scan resolution < (scale × desired resolution of an orthophotograph)/2, with the “scan resolution” (pixel size) in ȝm and the “desired resolution of an orthophotograph” in m. For example, if the orthophotograph is to resolve details of 0.5 m in size and the scale of the aerial photograph is 1 : 7500, then 1/7500 × 0.5 m = 66 ȝm. To adequately satisfying
Environmental Geology, 3 Remote Sensing
83
the rule of thumb, the pixel size for scanning should be 30 ȝm resulting in a square of four pixels that is 60 ȝm for a side. In order to obtain sufficient accuracy, the aerial photograph should be rectified by simple rectification, stereometric processing or differential rectification (Section 3.2.3). The decision for which rectification method to be used depends on the elevation differences in the area of interest. The limits for simple rectification are given in Table 3.2-1. For areas with greater topographic differences, differential rectification with an elevation model is the appropriate method to prepare precise orthophotographs and thematic maps. Rectification procedure Digital image processing is commonly used in photogrammetry. In its basic form of understanding, an input image is processed by a transformation function into an output image. The processing can involve changes in the geometric relationships and the radiometric content of the image. Radiometric transformations change the gray values of the pixels, e.g., in order to improve the spectral quality or contrast of the image. The geometric projection of an image onto a horizontal reference plane is called rectification or restitution. The spatial relationship between image coordinates and reference plan locations (e.g., the Earth's surface) is determined by ground control points. These points are used to rectify the output image into a topographic reference system. A difficult problem in the rectification is the determination of the transformation functions. Quadratic equations are suitable but require at least six control points to determine the equation coefficients. A location precision of less than one pixel can be obtained. Two approaches are commonly used to transform to original pixel values to the rectified image using these equations (ALBERTZ, 2001): x
Direct transformation: The location of each point in the input image is mapped to the output image and the value of the input pixel is assigned to the output pixel. Due to the geometric distortion of the input image this can lead to heterogeneous pixel occupancy in the output image. Interpolation to a regular pixel distribution is necessary after the transformation.
84 x
3.2 Photogrammetry
Indirect transformation: For each pixel of the output image, an inverse transformation function is used to determine the location of the corresponding pixel in the input image to obtain the desired pixel data. The coordinates of data points in the input image will not map to pixel centers. This procedure is commonly used.
In this context, different rules have been introduced to obtain “suitable” estimations of gray values for the pixel matrix of the output image (GÖPFERT, 1991; MATHER, 1999; ALBERTZ, 2001): x The nearest neighbor takes the pixel value of the input image closest to the calculated coordinate value. x Bilinear interpolation: The relevant pixel value is calculated by linear interpolation from the values of the four closest pixels around the calculated coordinate value. x Bi-cubic interpolation: interpolation using the 4 × 4 pixels neighborhood around the calculated coordinate value 2 . Before rectified aerial photographs can be merged to a photomosaic, the radiometric differences between the individual images in terms of brightness, contrast, and color are minimized by radiometric balancing. The available image processing software can be used for individual images or groups of images. Digital elevation models derived from stereo processing of aerial photographs A digital elevation model (DEM) can be produced by digital photogrammetric processing of overlapping aerial photographs or by other methods such as laser scanning or radar interferometry. The quality of the digital restitution of pairs of aerial photographs depends on the quality of the image correlation, as the most important part of stereoscopic restitution (FÖLLER & GERTLOFF, 1998). After scanning the aerial photograph and determination of the ground control points, the area of interest is divided into subareas with uniform terrain characteristics. The procedure for image correlation has to be selected for each subarea. Criteria for the selection of processing procedures are terrain texture, steepness of slopes, degree of urbanization, vegetation, terrain breaks, and raster width of the digitized image. For the automatic processing of a pair of aerial photographs, a raster cell size of one meter has proven to be suitable. Smaller raster sizes do not increase accuracy, while larger grid cells can lead to a loss of both detail and small land features. An alternative consists of closely spaced elevation measurements along terrain breaks and a more widely spaced grid in areas 2 More information for the gridding/interpolation algorithms is given in Part 6.
Environmental Geology, 3 Remote Sensing
85
with only minor elevation changes. Before calculation of elevation contour lines, errors caused by disturbing objects such as trucks, single trees and hedges and forested areas should be eliminated interactively. Digital elevation models derived from airborne laser scanning Laser scanning (Chapter 3.3) can be used to prepare highly accurate digital elevation models. These models can also be combined with other products of photogrammetry, such as orthophotographs, photomosaics, topographic maps and DEM from stereometric processing. The coordinates of the laser reflection points are calculated after data acquisition. These data are combined and compared with coordinate values from DGPS, the flight data, and the inertial navigation data from the aircraft. Disturbing objects, such as trees, buildings and cars, should be eliminated from the digital elevation model.
3.2.7 Quality Assurance On the commercial market, aerial photogrammetry is influenced by pricing pressures with possible negative impacts on the quality. On the other hand, photogrammetric products are increasingly used by organizations which do not have photogrammetric know-how. Therefore, quality control standards are a necessity. A number of companies have developed quality management systems. Different aspects of quality assurance for aerial photograph surveys are described in Section 3.1.7.
86
3.2 Photogrammetry
3.2.8 Personnel, Equipment, Time Needed Personnel
Equipment
Time needed
2-3 technicians or specialists
special survey aircraft with a flight management system, aerial metric camera, aerial film
1 day
film development
1 technician
roller transport processor
1 day
scanning of two aerial photographs
1 technician
photogrammetric scanner
1 hour
terrestrial GPS/DGPS measurements (ca. 10 ground control points)
1 driver 1 technician or specialist
1 4WD vehicle, 1 GPS or DGPS
1 - 3 days
preparation of orthophotographs or aerial photomosaics
1 specialist
PC, plotter, software
1 day
preparation of digital elevation model using aerial photographs: for 10 km² for 100 km²
1 specialist
PC, plotter, software
updating of existing map
1 specialist
PC, plotter, software
1 - 3 weeks
preparation of new topographic map
1 specialist
PC, plotter, software
4 - 8 weeks
aerial photography survey
5 - 10 days 4 - 6 weeks
3.2.9 Examples Preparation of orthophotographic base maps from aerial photographs and IKONOS data Up-to-date topographic maps (base maps) at a scale of 1 : 5000 or less are required for site investigations. Topographic maps at a scale of 1 : 50 000 published by the Royal Thai Survey Department are usually available for applications in Thailand and have proven to. However, these maps are insufficient as base maps for site investigations. Therefore, for the investigation of the area of the Dan Khun Thot landfill in the Nakhon Ratchasima province of Thailand, base maps had to be prepared from aerial photographs or high-resolution satellite images. IKONOS satellite images
Environmental Geology, 3 Remote Sensing
87
(UTM projection, zone 47N, 1 m pixel size, datum WGS84, panchromatic) and aerial photographs at a scale of 1 : 15 000 from the Royal Thai Survey Department were used. Cloud coverage of the IKONOS images was 1 % for site 1 and about 20 % for other sites. A ground check of the prepared maps with a handheld GPS revealed deviations in the coordinates for site 1 between 3 and 13 m. This accuracy is sufficient for site investigations. The proposed waste disposal site 1 (Fig. 3.2–5, red line) is located approximately 17 km southwest of Dan Khun Thot with an area of about 135 000 m2. The terrain is undulating with elevations between 235 and 265 m above sea level.
Fig. 3.2-5: Base map of the proposed Dan Khun Thot waste disposal site 1 (Thailand) prepared from IKONOS images from Nov. 27, 2002, coordinates: easting 0781750 to 0782750 and northing 1675000 to 1675750, Thai-Viet projection, UTM zone 47
88
3.2 Photogrammetry
Fig. 3.2-6: Orthophotograph of the proposed Dan Khun Thot waste disposal site 2 (Thailand) prepared from aerial photographs, coordinates: easting 0782600 to 0785200 and northing 16672080 to 1669100, Thai-Viet projection, UTM zone 47
Figure 3.2-6 shows an example of an orthophotograph prepared from aerial photographs at a scale of 1 : 15 000. Preparation of an aerial photomosaic and a topographic base map for site investigations CIR aerial photographs at 1 : 7500 scale were acquired to provide a base map for the Mae Hia waste disposal site near Chiang Mai in northern Thailand and for remote-sensing site investigations. The aerial photomosaic (Fig. 3.2-8) is the result of rectifying and geo-referencing individual aerial photographs (Fig. 3.2-7), merging and joining them. First, the CIR images were scanned with a 28 ȝm pixel size. A differential rectification was carried out considering elevation differences in parts of the investigation area. Contour lines from the official 1 : 50 000 topographic map and DGPS control points were used for this purpose. Thailand’s 1975 reference ellipsoid and the coordinate system of UTM zone 47 formed the geodetic reference system. The individual images and their joining were processed and merged as a mosaic.
Environmental Geology, 3 Remote Sensing
89
Fig 3.2-7: CIR-aerial photographs of the area around the Mae Hia waste disposal site in Chiang Mai, northern Thailand
Legend
² _____
control points by GPS topographic contour line
Fig 3.2-8: Rectified aerial photomosaic with contour lines and control points in the area of the Mae Hia waste disposal site
90
3.2 Photogrammetry
Fig 3.2-9: Aerial photographic map of the area around the Mae Hia waste disposal site
Fig 3.2-10: Topographic map of the area around the Mae Hia waste disposal site
Environmental Geology, 3 Remote Sensing
91
The aerial photographic map of Figure 3.2-9 was produced from the rectified and geo-referenced aerial photomosaic in Figure 3.2-8. On the basis of this aerial photographic map, a topographic base map (Fig. 3.2-10) at a scale of 1 : 5000 was prepared by on-screen digitizing and mapping using the GIS ArcView 3.2a interpretation software. This base map was used for the geological, geophysical, geochemical and remote-sensing site investigations of the Mae Hia waste disposal site. Only selected topographic elements, such as the perimeter of the disposal sites, streets, paths, rivers, canals, villages and towns, were plotted as a topographic base map on the maps with other geoscientific data. Development and updating of a digital elevation model of a landfill A digital stereo restitution of 1 : 5000 color aerial photographs was carried out to prepare a digital elevation model of a landfill. During digital stereo restitution, elevations were determined using a 15 m grid for most of study area and in more detail along terrain breaks (Fig. 3.2-11). Aerial photographs were taken annually and used to update the elevation model of the landfill. This is done to document changes in the surface of the landfill, to estimate the increase in volume of waste, and observation of shrinkage (compaction). Monitoring the relief and changes in volume informs the operating agency about the condition of the landfill over a period of time and about how long the site can be used before the limits are reached, as well as, supports planning efforts for new necessary sites.
Fig 3.2-11: Perspective view of a digital elevation model of a municipal landfill in Germany, Hansa Luftbild (German Air Survey), Münster
92
3.2 Photogrammetry
Preparation of a regional digital elevation model from airborne laser scanning Laser scanning observations can be used to monitor natural and manmade changes in relief. The area of investigation is located in a region of intense subrosion leading to caving to the surface and subsidence. Mining activities in the area have reinforced and intensified this process. A cone-shaped tailings and slag heap, rising about 150 m above the surrounding area, can be seen in the center of the derived digital elevation model (Fig. 3.2-12).
Fig 3.2-12: Regional digital terrain elevation contour model (system: Gauß-Krüger) of part of the Eisleben mining area in Germany from airborne laser scanning, BGR
Environmental Geology, 3 Remote Sensing
93
References and further reading ACKERMANN, F. (1995): Digitale Photogrammetrie – ein Paradigma-Sprung. Zeitschrift für Photogrammetrie und Fernerkundung, 63 (3), 106-115. ACKERMANN, F. (1996): Verfahrenstechnik und Potential des Airborne Laser Scanning. Hansa Luftbild Symposium 1996: Airborne Laser Scanning – Ein neues Verfahren zur berührungslosen Erfassung der Erdoberfläche, Münster, 25.04.1996. ALBERTZ, J. (2001): Einführung in die Fernerkundung, Grundlagen der Interpretation von Luft- und Satellitenbildern. Wissenschaftliche Buchgesellschaft, Darmstadt. ALBERTZ, J. & KREILING, W. (1989): Photogrammetrisches Taschenbuch – Photogrammetric Guide; Wichmann, Karlsruhe. BÄHR, H. P. & VÖGTLE, T. (1998): Digitale Bildverarbeitung. Wichmann, Heidelberg. BALTASAVIAS, P. & KAESER, C. (1999): Quality Evaluation of the DSW 200, DSW 300, SCAI and Ortho Vision Photogrammetric Scanners. Unpublished Report. BAUER , M. (1997): Vermessung und Ortung mit Satelliten. Wichmann, Karlsruhe. BRIESE, C. & PFEIFER, N. (2001): Airborne Laser Scanning and Derivation of Digital Terrain Models. In: GRÜN & KAHMEN (Eds.): Optical 3-D Measurement Techniques, V, Wichmann, Karlsruhe, 80-87. BURNSIDE, C. D. (1979): Mapping from Aerial Photographs. Granada, London. CICIARELLI, J. A. (1991): A practical guide to aerial photography: with an introduction to surveying. Van Nostrand Reinhold, New York. COLWELL, R. N. (1983): Manual of Remote Sensing. American Society for Photogrammetry, Falls Church (Virginia). ECKER, R., KALLIANY, R. & OTEPKA, G. (1993): High quality rectification and image enhancement techniques for digital orthophoto production. Photogrammetric Week 93, 142-155. EGELS, Y. & KASSER, M. (2001): Digital Photogrammetry. Taylor & Francis. FALKNER, E. (1995): Aerial Mapping: Methods and Applications. Lewis Publishers, Boca Raton. FÖLLER, J. & GERTLOFF, K. H. (1998): Überwachung der Verformung von Deponieoberflächen mit Methoden der digitalen Photogrammetrie. Zeitschrift für Vermessungswesen, 9, 287-294. FRIESS, P. (1996): Systemkonfiguration und Anwendungsbereiche des Airborne Laser Terrain Mapping – System ALTM 1020, Hansa Luftbild Symposium 1996: Airborne Laser. GIM International (2001): Product Survey on Airborne Laser Scanner - an Overview. GÖPFERT, W. (1991): Raumbezogene Informationssysteme. Wichmann, Karlsruhe. GUPTA, R. P. (2003): Remote Sensing Geology. Springer, Berlin.
94
3.2 Photogrammetry
HALLERT, B. (1968): Photogrammetry, Basic Principles and General Survey. Mc Graw Hill, New York. HEIPKE, C. (1999): Digital Photogrammetric Workstations, GIM International, 13, 1, 81. HILDEBRANDT, G. (1996): Fernerkundung und Luftbildmessung: für Forstwirtschaft, Vegetationskartierung und Landschaftsökologie. Wichmann, Karlsruhe HOFFMANN-WELLENHOF, B., LICHTENEGGER, H. & COLLINS, J. (2001): Global Positioning System. Theory and Practice, Springer, Wien. HOSS, H. (1997): Einsatz des Laserscanner-Verfahrens beim Aufbau des Digitalen Geländehöhenmodelles in Baden-Würtemberg. Mitteilungen d. Deutschen Vereins f. Vermessungswesen, Heft 1. KONECNY, G. (1997): From Photogrammetry to Spatial Information Systems. ITC Journal. KONECNY, G. & LEHMANN, G. (1984): Photogrammetry. de Gruyter, Berlin. KÖLBL, O. (1992): Popularisation of Photogrammetry. I. Arch. Ph. XXIX – B2, 601- 607. KRAUS, K. (1993): Photogrammetry. Vol. 1: Fundamentals and Standard Processes, Dümmler, Bonn. KRAUS, K. (1997): Photogrammetry. Vol. 2: Advanced Methods and Application. Dümmler, Bonn. KRAUS, K. (2000): Photogrammetrie. Vol 3: Topographische Informationssysteme. Dümmler, Köln. KUMM, W. (2000): GPS Global Positioning System. Delius, Klasing, Bielefeld. LILLESAND, T. M. & KIEFER, R. W. (1994): Remote Sensing and Image Interpretation. 3rd edition, Wiley & Sons, New York. LINDNER, W. (2003): Digital Photogrammetry – Theory and Applications. Springer, Berlin. LÜSCHER, H. (1944): Kartieren nach Luftbildern. Berlin. MATHER, P. M. (1999): Computer Processing of Remotely Sensed Images: An Introduction. Wiley & Sons, Chichester. MIKHAIL, E. M., BETHEL, J. S. & MCGLONE, J. C. (2001): Raumbezogene Informationssysteme, Indroduction to Modern Photogrammetry. Wiley & Sons, Chichester. REEVES, R. G. (1975): Manual of Remote Sensing. American Society for Photogrammetry, Falls Church (Virginia). RITCHIE, W., WOOD, M., WRIGHT, R. & TAIT, D. (1988): Surveying and Mapping for Field Scientists. Longman. SABINS, F. F. (1996): Remote Sensing: Principles and Interpretation. 3rd edition. Freeman and Co., New York. SCHNEIDER, S. (1974): Luftbild und Luftbildinterpretation. de Gruyter, Berlin. SCHWEBEL, R. (2001): Qualitätssicherung für Bildflug und analoges Luftbild durch neue DIN-Norm, Photogrammetrie, Fernerkundung. Geoinformation, 1, 39-44.
Environmental Geology, 3 Remote Sensing
95
STUTTARD, M. & DOWMAN, I. (1998): Guidelines for Quality Checking of Geometrically Corrected Remotely Sensed Imagery, Unpublished Report. SCHWIDEFSKY, K. & ACKERMANN, F. (1976): Photogrammetrie – Grundlagen, Verfahren, Anwendungen. Teubner, Stuttgart. SEEBER, G. (1989): Satellitengeodesie. de Gruyter, Berlin. STEINER, T. D. & MERRIT, P. H. eds. (1998): Airborne Laser Advanced Technology, 338, S P I E- International Society for Optical Engineering. TEUNISSEN, P. J. G. & KLEUSBERG, A. (1998): GPS for Geodesie. Springer, Berlin. VEGT, J. W. & HOFFMANN, A. (2001): Airborne Laser Scanning Reaches Maturity. Geoinformatics Sept. 2001, Emmelord, 32-39. WOLF, P. R. & WITT, B. (2000): Elements of Photogrammetry. Mc Graw Hill, New York.
Environmental Geology, 3 Remote Sensing
97
3.3 Nonphotographic Imaging from Aircraft and Space-borne Platforms FRIEDRICH KÜHN, BERNHARD HÖRIG & DIETMAR SCHMIDT, with a contribution by TILMANN BUCHER
3.3.1 Principle of the Methods Instead of using light-sensitive film, nonphotographic imaging systems detect the incoming electromagnetic (EM) radiation with semiconductor detectors or special antennas. While photography is limited to a spectral range from 0.3 to 0.9 µm, multispectral scanners (MSS) can operate in wavelength regions from 0.3 to approximately 14 µm (LILLESAND & KIEFER, 1994; GUPTA, 2003). This range includes radiation in the near ultraviolet, visible light, near, middle, and thermal infrared (Table 3.1-1). MSS can work in rather narrow spectral bands of a few nanometers. Thus, such remote sensing can focus on specific spectral features to identify detailed physical and chemical characteristics of the land surface objects of interest. The basic operating characteristics and image geometry of multispectral and thermal scanners are similar. Both passive and active methods are used for nonphotographic imaging. Passive methods sense the natural electromagnetic radiation reflected or emitted from the Earth’s surface. Active sensors contain their own source of EM radiation (e.g., laser or radar). They transmit a coherent signal and receive the reflected radiation to obtain geometric, geological, and environmental information from the Earth’s surface. The most important nonphotographic imaging systems for geoscientific site investigations are: x opto-mechanical line scanners or whisk broom scanners, x opto-electronic line scanners, using charge coupled devices (CCD) basic sensor unit and are also refered to as linear scanners or push broom scanners, x digital cameras (CCD array), x imaging spectrometers, x microwave sensors, mainly radar-based systems, and x laser scanners. They are operated from aerial platforms and from space-borne platforms. During last decades, the technical development of nonphotographic imaging sensors has been and continues to be very rapid. The theoretical background and technical principles of nonphotographic remote-sensing systems are
98
3.3 Nonphotographic Imaging
described in more detail in COLWELL (1983), KRONBERG (1985), GUPTA (1991 and 2003), ALBERTZ (2001), and SABINS (1996). Technically, space-borne imaging systems operate similarly airborne systems. Differences arise from the longer distance between the sensor and the object on the Earth surface (affecting spatial resolution and observation noise), the methods of data transfer, the temporal resolution, and costs for operation and data. There are some limitations on the use of space-borne data. Because of the greater distance there is more atmospheric interference during image capture. This can result in more radiometric distortions in the image data, e.g., object boundaries can be blurred compared to aerial photos with similar spatial resolution (KERSTEN et al., 2000). For geologic applications, data acquisition is sometimes restricted to a particular season and the study area must be free of clouds. Thus, the chances of adequate data capture are somewhat reduced, because the Landsat Thematic Mapper, for example, covers the same site on the Earth only every 16 days, which may not coincide with optimum weather conditions. Bad weather results in a delay in the acquisition of the data or makes the use of archival material necessary. Satellite images and aerial photographs can be interpreted using the similar analytical methods. Sophisticated processing and interpretation of nonphotographic data require special advanced software and appropriate skills of the human analysist. Satellite images differ from aerial photographs in that they normally have a lower spatial resolution with larger area coverage. The first Landsat sensors in the 1970s had pixel resolutions of about 80 by 80 m. The Landsat-TM sensors of the 1980s produced images with a spatial resolution of 30 by 30 m, and the panchromatic band of the Landsat-ETM+ (launched in 1999) even provides 15 by 15 m. Similar to aerial photographs, overlapping satellite images can be analyzed stereoscopically. Data from nonphotographic systems (Landsat Thematic Mapper, SPOT, ERS-1, ERS-2, RADARSAT, IKONOS 2, etc.) may be obtained in digital form on magnetic tape, optical disk, CD-ROM or as hard copy prepared according to standardized image processing. Nonphotographic remote-sensing systems have the following basic advantages over conventional aerial photographic cameras: x the ability to detect reflected or emitted radiation from objects on the Earth's surface over a wider wavelength range (near-infrared, thermal radiation, microwaves, e.g., radar), x the ability to measure electromagnetic (EM) radiation in narrow bands simultaneously with the same optical system, x higher radiometric and spectral resolution, x the ability to directly record data in digitized form supporting more rapid data processing, x electronically generated data are more amenable to calibration, and
Environmental Geology, 3 Remote Sensing
99
x the ability to transmit data from the sensor to a ground-based receiving station providing immediate availability of the data. Images from satellite-based remote-sensing systems are primarily used to obtain data for large areas. For a long time, the spatial resolution of a satellitebased remote-sensing system was usually insufficient to delineate and identify small objects on the Earth’s surface. During the last several years, new satellite-based remote-sensing systems, e.g., IKONOS 2 and QUICKBIRD-3, have improved spatial resolution comparable to aerial photography. These high-resolution commercial satellite images can be used to generate or update topographical maps for a site of interest and for investigation of the region around the site being investigated. After rectification using a digital elevation model (DEM) and ground control points, IKONOS and QUICKBIRD data can be used to produce orthorectified images with a geometric accuracy of 1 - 5 m. However, airborne sensors with spatial resolutions of several centimeters are still more suitable for detailed investigations, e.g., the evaluation of the vitality of trees or the search for leachate water, springs, ground destabilization, thermal features. For multi-temporal analysis, aerial photographs are still indispensable (Chapter 3.1) due to the extensive historical archives. To obtain most appropriate information for a specific purpose, it is recommended to combine photographic and nonphotographic data and analyze them synergistically.
3.3.2 Applications x Analysis of the development of waste disposal, mining, and industrial sites and their surroundings, x preparation and/or updating of orthophotographs and topographic maps, x preparation of digital elevation models, x evaluation and assessment of environmental and geological conditions using methods similar to aerial photograph interpretations, x assessment of previous geological and environmental conditions at a waste disposal, mining, and industrial site, x search for seepage of water at the edges of landfills and heaps, x localizing seepage, springs, and soil moisture anomalies, x investigation of natural and artificial drainage systems, x localizing high and low permeability areas for water, x recognition and monitoring of destabilization processes (horizontal and vertical land movement),
100
3.3 Nonphotographic Imaging
x analysis of the vitality of vegetation, x assessment of surface water conditions, x delineation of fractures and lineaments, x inventory of sites suspected to be contaminated, x mapping of land use patterns and biotopes, x search for new waste disposal sites, x localizing heat sources inside landfills (“hot spots”), x detection of gas emission and surveillance of gas emission ducts in landfills, x assessment of mining disposals and associated weathering products in mining districts, and x detection of hydrocarbons or materials containing hydrocarbons on the ground at oil fields, refineries, chemical factories, petrol depots, airports, coking plants, gas works, along or close to transport ways and near waste disposal sites.
3.3.3 Fundamentals The term “image” is used as a general term for a picture representation of a scene or object recorded by a remote-sensing system. It is usually restricted to representations acquired by nonphotographic methods. Nonphotographic imaging systems consist of an optical component, a detector, a data recording system, and in space-borne applications a telemetric system for transmitting the data to the ground base stations. The optical component contains lenses and mirrors to collect the incoming radiation, and filters, prisms, and/or gratings to split the signal into different wavelengths. The detector includes devices transforming optical energy into electrical signals, such as photoemissive detectors, photo-conductive detectors, and photodiodes. The frequently used charge-coupled devices (CCD) are photodiodes. The efficiency of nonphotographic imaging systems and the possible applications depend on the following basic parameters and system-related conditions: The quality of images depends on the aperture, the opening in a remote-sensing system that admits electromagnetic radiation to the film or detector, and the instantaneous field of view (IFOV), the solid angle in which the detector is sensitive to radiation. The IFOV is normally expressed as the cone angle E within which the incident energy is focused on the detector. The angle E is determined by the optical component and the size of detectors and is measured in radians. Spatial or ground resolution is directly related to the IFOV and describes the ground area sensed by one of the sensor/detector
Environmental Geology, 3 Remote Sensing
101
elements and determines smallest terrain features to be recognized in the resulting images. All energy (from the land surface) arriving at the detector within the IFOV contributes to the detector response at any instant. A larger IFOV results in more energy being caught by the detector. This improves the ability of sensor systems to discriminate gentle differences in radiation intensity (radiometric resolution). The smaller the IFOV gets, the smaller is the size of the minimum resolvable scene element and the finer the ground resolution, but with the less radiation arriving at the sensor. Thus, there is a trade-off between spatial and radiometric resolution. A similar trade-off exists for width of spectral bands. In general, the spectral resolution is the ability of sensor systems to discriminate spectral differences and is determined by the number of spectral bands, their bandwidths, and the spectral region covered. The amount of radiation arriving at the sensor decreases for narrower spectral bands. Given a constant system noise, a lower amount of acquired radiation signal causes the signal-to-noise ratio to decrease and ultimately lowers abilities to analyze the spectral signal. The signal-to-noise ratio is an essential parameter for characterizing the capabilities of imaging spectrometers. The dwell time is the time required for a detector IFOV to sweep across a ground resolution cell. The angular field of view is the angle subtended by lines from a remote-sensing system to the outer margins of the strip of terrain that is imaged by the system. The width of the strip of terrain that is imaged by a scanner system – called ground swath – is determined by the total angular field-of-view (FOV). Depending on the nonphotographic system, the images show distortions, i.e. changes in shape and position of objects with respect to their true shape and position on the Earth’s surface. In general opto-electronic scanners have images with more simple geometric properties than opto-mechanical line scanners. Opto-mechanical line scanner images have a panoramic distortion. The constant angular velocity of the rotating scanner mirror results in a scale distortion perpendicular to the flight direction. The width of ground resolution cells (IFOV) increases with the lateral distance from the nadir to the edge of the swath. The image scale in the direction of flight is constant. Images taken by optical-electronic scanners have no panoramic distortion. In optomechanical line scanner images vertical features are displaced at right angles from the nadir line (one-dimensional relief displacement). Flight parameter distortions have to be corrected for both scanner types, particularly in the processing of images from airborne systems. Opto-mechanical line scanners Opto-mechanical line scanners sense the Earth's surface by scanning across a swath of land. This scanner system is often referred to as a "whisk broom" scanner or “across-track” scanner. Opto-mechanical line scanners have been
102
3.3 Nonphotographic Imaging
frequently used in the last several decades because they can be operated multispectrally at a wide range of wavelengths from 0.3 to 14 µm. Optomechanical line scanners are used for both airborne applications, e.g., the Daedalus-AADS scanner (Fig. 3.3-2), and space-borne applications, e.g., MSS, TM, and ETM+ scanners on Landsat, MTI and ASTER (TIR part) satellites (Section 3.3.4). The principle of this type of scanner – consisting of opto-mechanical line scanning system, detector unit, and data recording system – is shown in Figure 3.3-1. It is similar for both aerial and space-borne applications. A moving mirror tilted at 45° is the basic element of an opticalmechanical line scanning system. It scans the area by rotating or oscillating perpendicular to the flight direction. Due to the forward motion of the sensorcraft a two-dimensional data array (image) is obtained. The size of the smallest discrete area indentifyable in an image (pixel) depends on the flight altitude and the IFOV. The spatial resolution of most scanners is between 1 and 2.5 mrad. A scanner with an IFOV of 1.5 mrad, equivalent to 0.086°, will scan with a pixel size of 1.5 m u 1.5 m at a flight altitude (hg) of 1000 m. These figures are for the center of a scanned swath. As a rule of thumb, to be recognizable on a scanned image an object on the Earth’s surface should have a minimum size about 2.8 times the pixel size (ALBERTZ, 2001). For the preceding example (pixel: 1.5 m u 1.5 m; hg = 1000 m), an object would be at least 4.2 m across to be clearly identifiable. Airborne opto-mechanical line scanners are operated at flight altitudes between 300 and 12 000 m.
Fig. 3.3-1: Principle of opto-mechanical line scanners, modified after GUPTA (1991)
Environmental Geology, 3 Remote Sensing
103
There often is a compromise between spatial resolution and swath width that has to be considered for mission planning. The margins of the scanned image usually show distortion of the pixels (panoramic distortion). They have to be geometrically corrected later. Because of inherent aircraft instability, geometric rectification of aircraft data is a major problem if the data needs to be georeferenced. Also, specific areas on the edges of the swath may be not acquired. With satellite-borne scanners, these distortions are usually irrelvant. A scan line is the result of a single scan (moving mirror) across the width of the swath. The length of one scan line, i.e., the width of the swath, is defined by the angular field of view (FOV) of the scanner. This angle is usually between 90° and 120°. Due to the forward motion of the aircraft (or satellite), the terrain is scanned line by line resulting in a continuous image. Only if the synchronization is not properly adjusted overlaps or gaps will appear between the scan lines. The acquisition of high-quality images requires a perfect synchronization between the scan frequency of the mirror and the speed of the aircraft or satellite.
Fig. 3.3-2: Daedalus-AADS scanner image of the Münchehagen waste disposal site in Lower Saxony, Germany. The image was taken on July 6, 1989, by DLR Oberpfaffenhofen for BGR (channel 3 = red, 4 = green, and 5 = blue).
104
3.3 Nonphotographic Imaging
The reflected or emitted signal from the Earth surface reaches the scan mirror and passes through the optical system to the detector unit. The detectors transform the optical signals into electrical signals. The intensity of the electric signal largely depends on the intensity of the incoming radiation and, thus, on the spectral properties (absorption, reflection, or emission) of the ground surface. Multispectral scanners with hundreds of spectral channels have been developed and applied. The radiation is divided by a system of gratings and prisms into partial spectra and then passed to specialized detectors. Usually, it is necessary to use several different detectors because different parts of the electromagnetic spectrum between the visible part of spectrum and thermal radiation require different semiconductor crystals. It is common to employ Si detectors for the visible range and NIR-I and PbS detectors for the NIR-II part of the electromagnetic spectrum. InSb detectors for thermal radiation in the mid-infrared (MIR-II) range require cooling with liquid nitrogen (77 K). A temperature resolution on the order of 0.1 qC is attainable (LILLESAND & KIEFER, 1994). The electrical signals from the detectors, commonly measured as voltage, are stored digitally in the data recording system. In addition to the relatively low ground resolution, the major disadvantage of optical-mechanical line scanners is their sensitivity to mechanical failure. Numerous moving parts must be adjusted very precisely and are subject to aging and wear. In addition, previously mentioned distortions related to movement of the aircraft and panoramic distortion at the edges of the scanned swath, require computer correction. Related processing and interpretation of the data requires access and capabilities for operating a digital image processing system. To obtain optimum results, a geologic interpreter should be involved in the processing of the data so they are aware what processing has been done with the original data. Geographical and geological knowledge about the target study area and a well-focused, goal-orientated approach are prerequisites to a successful study. A more detailed explanation of the construction and performance of opto-mechanical scanners can be found in KRONBERG (1985), GUPTA (1991 and 2003), LILLESAND & KIEFER (1994), SABINS (1996) or ALBERTZ (2001). Although being technologically outdated by opto-electronic scanners, the primary reason opto-mechanical line scanners are still in use is their ability to record thermal radiation. Furthermore, such sensors are quite effective including a larger number of spectral bands. In the course of technological development, however, it is assumed that future generations of opto-electronic scanners will be capable of routine data acquisition in the thermal wavelength ranges.
Environmental Geology, 3 Remote Sensing
105
Information on the composition of surface materials and identification of temperature anomalies in and around a landfill can be obtained from thermal scanner images. The detected thermal properties of an object are determined both by the material of the object and by the character of the surrounding environment. The intensity of thermal radiation is dependent on both the surface temperature and the emissivity of the materials. The following physical and environmental parameters influence the thermal behavior of an object at the Earth’s surface (after KRONBERG, 1985): Physical parameters x color x composition x surface properties x density x porosity or pore volume x permeability x water content
Environmental parameters x topographical position in the field x orientation of the surface with respect to the sun x meteorological conditions x microclimate x humidity x time of day, season x type and extent of vegetation cover
Remotely sensed radiation temperatures have to be calibrated against temperatures measured at the ground surface. Thus, calculation of surface temperatures requires considerable field work; an effort only necessary for specific applications. Any given object in a thermal image can be identified as “warm” or “cold” relative to its surroundings (Fig. 3.3-3). Environmental influences can cause confusing overlap or even mask thermal radiation emitted from natural or artificial objects, creating false anomalies in the thermal images. The diurnal temperature cycles of two very different materials show that it is impossible to generalize about “warm” or “cold” characters of materials on the basis of their physical characteristics. According to GEBHARDT (1981), physical and environmental factors may act independently of each other or they may be synergetic or antagonistic to each other. In addition, the environmental factors are difficult or impossible to determine only on the basis of the images. A landfill, mining or industrial site usually covers an area smaller than 1 km2. Thus, highly sophisticated scanning systems designed for thermal mapping of larger areas may not be economically feasable in every case. Simple thermographic images often fit the requirements of site characterization best (Fig. 3.3-3, Tables 3.3-2 and 3.3-3).
106
3.3 Nonphotographic Imaging
Fig. 3.3-3: Thermal image of the western rim of the Schöneicher Plan landfill and the adjacent area. The margin of the deposit (right) is warmer than the field to the left. Relief, groups of trees, and local winds generate exogenic thermal anomalies. Temperatures shown in °C (image taken on May 13, 1993, at 05:19 CET by F. BÖKER and F. KÜHN, BGR).
Optical-electronic scanners Optical-electronic scanners are often referred to as CCD (Charge Coupled Device) linear array scanners, push-broom scanners, or along-track scanners. In principle, they are similar to photographic cameras. Instead of film, CCD arrays are installed in the focal plane. CCD arrays consist of a large number of detector elements, e.g., silicon-based photo detectors. Several thousand detectors may be placed on a single 1.5 cm2 chip. Each individual spectral band, or channel, requires its own linear CCD array. This usually limits the number of spectral bands to be put on one sensor. The CCD chips are mounted in the focal plane of an optical electronic scanner. The optical system focuses the incoming radiation on the surface of the light-sensitive chips. The radiation on each detector is integrated over a short time interval (dwell time) and transformed into electrical signals and recorded. The intensity of each signal depends on the spectral response of the related landscape feature. The detector chips are in a line perpendicular to the direction of flight. The swath is scanned due to the forward motion of the sensorcraft line by line in the manner a push-broom is used (Fig. 3.3-4). Optoelectronic scanners are operated from airborne platforms as well as from space.
Environmental Geology, 3 Remote Sensing
107
Fig. 3.3-4: Principle of an optical-electronic scanner (CCD linear array scanner, often called a push broom scanner), modified after GUPTA (1991)
The size of the ground resolution element of an opto-electronic scanner depends on the size of the detector cell as projected through the optics onto the ground surface. The ground resolution is, thus, dependent on the flight altitude hg, the dwell time W of each scan, and the velocity v of the aircraft. The ground resolution element is defined by a line L1 parallel to the direction of flight (Equation 3.3.1) and L2 perpendicular to this direction (Equation 3.3.2) as follows: L1 L2
vW a hg cf
(3.3.1) ,
(3.3.2)
where a is the distance between the centers of adjacent detector elements on the chip and cf is the focal length of the optical system. The major advantages of opto-electronic scanners compared to optomechanical scanners are their better reliability (due to the lack of moving parts), higher geometric precision along a scan line, and a longer durability. A longer dwell time enables a much stronger signal with better radiometric resolution and poorer spatial resolution. Disadvantage of this scanner type is the need to calibrate many more detectors. It is still difficult to produce chips with rows of detectors that are sensitive in the thermal infrared, are close enough together and can be individually cooled during flight. But, it is only a matter of time until the technology extends detection capability beyond the range of visible light (VIS) and parts of the near infrared (NIR-I and II) spectrum.
108
3.3 Nonphotographic Imaging
In multispectral opto-electronic scanners, the incoming radiation is splited into its spectral components by a grating, prism or color filters. It is possible to scan the surface over a selected wavelength interval by arranging several rows of detectors. The CASI scanner (described in the following section) is a multispectral recording system which exclusively uses an opto-electronic scanning system. A more detailed explanation of the construction and performance of optoelectronic scanners can be found in GUPTA (1991 and 2003), LILLESAND & KIEFER (1994), SABINS (1996) or ALBERTZ (2001). Digital cameras The principle of remote-sensing systems known as digital cameras is quite similar to CCD linear array scanners (push-broom scanners) except that a twodimensional array (matrix) is used instead of a linear array. At present array sizes of 4 k × 4 k to 9 k × 9 k are common. Digital cameras with 20 k × 20 k CCD matrix sensor chips will be available soon. It is expected that in the near future digital cameras will be used for most topographic mapping surveys. It is easier to develop digital elevation models from digital camera images than from aerial photographs. The spatial resolution of digital cameras depends on the number and size of the sensor elements on the chip, flight altitude, and optical sensor parameters (e.g., focal length and aperture). The spectral sensitivity of silicon sensor chips limits digital cameras to the spectral range 0.4 – 1.0 µm. Space-borne multispectral systems usually provide images in the four spectral bands: blue, green, red and near-IR (GUPTA, 2003). The high resolution stereoscopic camera HRSC (Table 3.3-12) has the capabilities of both CCD linear array scanners and digital cameras. The HRSC contains nine CCD lines in the focal plane of the camera. Five lines are used to acquire panchromatic stereoscopic-images from different view angles and four lines are used to obtain multispectral data. Further examples of digital cameras are the Digital Modular Camera DMC (Table 3.3-13) and the ADS 40. They are suitable for photogrammetric surveys because their accuracy in the horizontal and vertical planes is of the order of centimeters, similar to a “metric” camera. Besides for airborne missions, CCD array technology is used on space platforms such as IKONOS, QuickBird, OrbView and ISI-EROS (GUPTA, 2003). An example of an IKONOS image is given in Section 3.2.9. Imaging spectrometers Airborne Imaging Spectrometers (AIS), also called hyperspectral scanners, operate on the principle of the push broom or whisk broom. The term hyperspectral is used to indicate the large number of contiguous spectral bands. Whereas opto-mechanical and opto-electronical scanners commonly
Environmental Geology, 3 Remote Sensing
109
used for multispectral sensing have 4 - 10 spectral channels with bandwidths of 100 - 200 nm, hyperspectral scanners record images of the Earth’s surface in 50 - 300 narrow bands with bandwidths between 1 - 20 nm in the visible and near-infrared parts of the electromagnetic spectrum. Thus, a quasicontinuous spectral signature for each pixel or ground resolution cell can be obtained (GUPTA 1991 and 2003, SABINS 1996). Radiation-splitting optics, CCD linear, and two- and multi-dimensional detector arrays are used to measure the radiation intensity as described above. Imaging spectrometry is based on reflectance spectroscopy – the study of the reflection, absorption and scattering of the solar radiation using laboratory and/or field spectrometers. The spectral features of minerals, rocks, vegetation types, manmade materials, water and dissolved organic compounds and environmental contaminants are recorded in digital spectral libraries. If these materials have unique and identifiable absorption features, they can be mapped by analysis of the imaging spectroscopy data. The wavelength region of interest is the portion of the electromagnetic spectrum in solar reflection band from 0.4 to 2.5 µm. Airborne imaging spectrometers provide data with a ground resolution from 0.5 - 20 m. In general, pixels may represent a spectral mixed signal of various materials inside this elementary mapping area. This limits the ability to clearly identify specific materials on the basis of the pixel spectra, e.g., comparison with standard spectra from a reference library. The same is true for classification of the pixels to map materials from spectral data. However, there are a suite of image analysis techniques focused on sub-pixel spectral characterization of the material compositions, i.e. spectral unmixing or matched filter analysis. In the 1990s, several German remote-sensing companies have explored the Canadian Compact Airborne Spectrographic Imager CASI. This sensor measures VIS to the NIR signals, i.e. 0.43 ȝm - 0.87 ȝm (Fig. 3.3-5). Theoretically, CASI contains 288 spectral channels, each with a bandwidth of 1.8 nm. Due to the large data volumes, not all of the 288 channels were used for data acquisition. The German Aerospace Center (DLR, Deutsches Zentrum für Luft- und Raumfahrt) has been developing and applying a digital airborne imaging spectrometer (GER-DAIS-7915) since 1995. NASA's hyperspectral AVIRIS scanner (Airborne Visible InfraRed Imaging Spectrometer) has been used mainly by U.S. agencies and companies to investigate mining districts in North America and to evaluate the hazards of the mining waste to ground and surface water (PETERS & PHOEBE, 2000; KING et al., 2000).
110
3.3 Nonphotographic Imaging
Fig. 3.3-5: CASI image taken on June 26, 1993, showing the SW corner of the Schöneiche landfill (right) and adjacent area. The image shows the vegetation index NDVI = (NIR - red) / (NIR + red) obtained using one recording channel each in the red and near infrared (NIR-I). The green/blue color of the zone in the center indicates poor growth on soil above permeable near-surface beds. (Images and image processing: WIB GmbH, Berlin, for BGR)
The Australian HyMap™ (Hyperspectral Mapping) system and its predecessor model Probe-1 (Table 3.3-16) is known for its outstanding signal-to-noiseratio of about 500:1, hence it provides clear and reliable spectral information. HÖRIG et al. (2001) and ELLIS et al. (2001) detected hydrocarbons and materials containing hydrocarbons on the ground using a HyMap (Table 3.3-16) or Probe-1. The Ekwan system (Table 3.3-17) is a new instrument with an even higher spatial and spectral resolution. Only the Modular Optoelectronic Scanner MOS of the German Aerospace Center (DLR) is operated on a satellite (ALBERTZ, 2001). Imaging spectrometers are sophisticated and rather expensive instruments. They are usually produced as individual instruments or in small numbers by technical laboratories or specialized companies. Because of the rapid technological development of remote-sensing devices, the number of manufacturers and new instruments has increased considerably. A summary of current and planned imaging spectrometers is given in Table 3.3-18. The cost of acquisition and operation of modern and technically advanced imaging data remains high. If possible, cost and data should be shared among different users or agencies.
Environmental Geology, 3 Remote Sensing
111
Microwave remote-sensing methods, mainly radar-based Active and passive remote-sensing systems have been used for measuring microwave electromagnetic radiation in wavelengths of 1 mm to 1.0 m (Table 3.1.1). Passive microwave sensor systems (radiometers) are commonly used to measure the intensity of natural radiation in the wavelength range 1 mm to 30 cm for mapping purposes (GUPTA, 2003). The signals recorded by passive microwave sensor systems represent surface temperature, emissivity, electrical and structural properties of the ground objects. Therefore, microwave radiometers can be used for geological and environmental investigations. However, the rather small amounts of natural spectral emissions in this spectral range only allow for coarse spatial resolution data to be acquired.
Fig. 3.3-6: Principle of side-looking airborne radar (SLAR), GUPTA (2003), (a) terrain illuminated with one transmitted radar impulse, (b) received back-scattered signal from the ground as an amplitude-time function
112
3.3 Nonphotographic Imaging
Most microwave remote sensing activities have been focused on active microwave systems particularly radar-based methods. Radar (acronym for Radio Detection and Ranging) illuminates an area with its self-produced coherent radiation in the cm-wavelength range. The long-wave electromagnetic radiation of radar penetrates clouds and is independent of natural radiation, e.g., can also be used at night. This all-weather and any-time capability is one of the distinct advantages of radar. Due to the oblique illumination angle of the transmitted radar signal to the Earth’s surface (Fig. 3.3-6), radar images are especially useful for detecting slight changes in topography due to the radar shadowing effect. This property is useful for mapping topographically expressed faults. Radar measurements are also sensitive to differences in soil moisture content. Common disadvantages of radar data are the complexity of data acquisition, geometric distortions, usually lower spatial resolution than optical systems and challenges in data interpretation. Figure 3.3-6 shows the principle of side-looking airborne radar (SLAR). In SLAR systems, a transmitter and receiver is mounted on an aircraft platform. A microwave pulse generated by the transmitter is obliquely send from the antenna to the Earth's surface illuminating narrow strips on the ground. The antenna is mounted on the side of the fuselage with a radiation direction perpendicular to the azimuth direction or flight direction and with a given look angle (look angle – angle between the line to the nadir from the antenna and the transmitted ray, depression angle – compliment of the look angle). The back-scattered signals (radar echoes) are acquired by the receiver antenna and converted to an amplitude-time function. Echoes from points on the ground closer to the antenna are recorded earlier as those further away. After the echo from the far range of the swath is received the next radar impulse is emitted. Strip by strip, the ground is scanned along the flight line, and a radar image is generated. In common radar systems the same antenna is alternately operated as transmitter and receiver antenna. The ground resolution of SLAR images is different in the range direction and the flight direction (azimuth). The range resolution is related to the duration of the transmitted radar pulse and to the look angle. Short pulses and a smaller look angle improve the range resolution. The resolution in the direction of flight (azimuth resolution) depends on the antenna characteristics and, therefore, on antenna length and on the wavelength of the transmitted radar wave. Shorter wavelengths and a longer antenna improve the resolution in the direction of flight. This resolution decreases from the near range to the far range. The possible antenna length is limited. With respect to azimuth resolution, radar systems are subdivided into real-aperture radar (RAR) and synthetic-aperture radar (SAR). In RAR systems, the azimuth resolution is determined mainly by the physical length of the antenna. In SAR systems, the azimuth resolution is determined by processing Doppler shift data for multiple return pulses from the same object to create synthetically longer antenna. Consecutive antenna
113
Environmental Geology, 3 Remote Sensing
positions are treated as if they were elements of one large antenna. This results in a better azimuth resolution than obtained with RAR systems without physically changing the length of the antenna. Sophisticated algorithms are necessary for SAR data processing. Current aerial and space-borne systems use SAR technology. Besides ground resolution, radar wavelength, beam polarization, the most important parameters for SAR systems are look angle and swath width. Table 3.3-1 gives the most frequently used radar bands. Microwaves with shorter wavelengths are more attenuated in the atmosphere. The penetration depth and the interaction of radar waves with materials on the Earth's surface are strongly influenced by the wavelength and polarization of the system. The backscatter signal depends on the dielectric properties (e.g., soil/plant moisture) and geometric characteristics such as roughness and shape of Earth surface types. Table 3.3-1: Radar bands used in SLAR systems after GUPTA (2003); values in parentheses indicate the commonly used radar wavelengths Radar band Ka Ks Ku X C S L P
Wavelength in cm 0.8 - 1.1 (0.86) 1.1 - 1.7 1.7 - 2.4 2.4 - 3.8 (3.1) 3.8 - 7.5 (5.7) 7.5 - 15.0 (15) 15.0 - 30.0 (23.5) 30.0 - 100.0 (50)
Frequency in GHz (109 cycles s-1) 40.0 - 26.5 26.5 - 18.0 18.0 - 12.5 12.5 - 8.0 8.0 - 4.0 4.0 - 2.0 2.0 - 1.0 1.0 - 0.3
The radar response signal is a complex number consisting of a real and an imaginary part. The amplitude and the phase of the signal can be calculated from its real and imaginary parts, respectively. Whereas conventional SAR imaging usually only use the amplitudes of the radar signals, interferometric SAR (InSAR) and polarimetric SAR (PolSAR) utilizes the phase measurements as well. The InSAR method does not use the phase information of single image for the interferograms but the phases of two images of the same ground scene recorded by SAR antennas at different locations and/or at different times. Detailed digital elevation models (DEM) can be derived from the phase differences of the interferograms. Differential interferometric SAR uses interferograms acquired over the same area at different times to detect displacements of the Earth’s surface in the centimeter range due to earthquakes, volcanic events, landslides or land subsidence. Additional information about the use of microwave remote sensing can be found in the literature (COLWELL, 1983; TREVETT, 1983; KRONBERG, 1985; WOODING, 1988; GUPTA, 1991 and 2003; SABINS, 1996).
114
3.3 Nonphotographic Imaging
Laser scanning Airborne laser scanning is an active remote sensing system as well. A pulsed laser beam scans the Earth’s surface in a strip across the flight line (Fig. 3.3-7). The pulse frequency and the speed of the aircraft determine the measured point density and the resulting mesh size of the data grid. An optomechanical scanner with rotating mirror guides the laser beam across the flight path. The path of the reflection points is a zigzag line due to the forward movement of the aircraft and the sideward oscillation of the scanner’s mirror. Complicated data processing derives the coordinates (position and elevation) of each terrain point. Data collection is possible even during the night and under cloudy conditions. The current standard for image accuracy is about 15 cm horizontally and 5 - 50 cm vertically for a flying altitude of 1000 meters above ground during data acquisition (VEGT & HOFFMANN, 2001). A single laser signal can be reflected before it reaches the Earth’s surface, i.e. by vegetation such as grass, trees, and shrubs. Therefore, high-resolution surveys, e.g., for monitoring of land subsidence, should be carried out in spring before the vegetation begins to grow. Suitable ground control points should be selected on the basis of the survey objectives. The last step is a classification of the terrain characteristics (topography, vegetation, buildings, etc.).
Fig. 3.3-7: Principle of data acquisition with an optical-mechanical laser scanner with a rotating mirror
Environmental Geology, 3 Remote Sensing
115
Modern navigation technology, such as DGPS and INS (Inertial Navigation System), are used in order to estimate the precise aircraft position. Irregular movements of the aircraft (dipping, rolling, tilting) during the flight are recorded and stored in the INS unit. This information is used to calculate the coordinates of points in the image data along the flight path. The use of INS and DGPS guarantees a position and altitude precision in the order of 5 - 20 cm. An example of airborne laser scanning is given in Fig. 3.2-11.
3.3.4 Instruments The specifications of selected airborne and space-borne nonphotographic remote-sensing systems are given in tabular form for the following categories: x thermal scanners, Tables 3.3-2, 3.3-3 x satellite multispectral imaging systems, Tables 3.3-4, 3.3-5, 3.3-6, 3.3-7, 3.3-8, 3.3-9 x space-borne radar systems, Tables 3.3-10, 3.3-11 x airborne digital cameras, Tables 3.3-12, 3.3-13 x imaging spectrometers and hyperspectral scanners, Tables 3.3-14, 3.3-15, 3.3-16, 3.3-17, 3.3-18 and x airborne laser scanners, Table 3.3-19. Field spectrometers operating in the spectral range 0.4 - 2.5 µm are used to provide basic information about the spectral properties of materials and to calibrate and validate remote sensing data. Airborne nonphotographic imaging systems include – besides the imaging sensor(s) – a navigation system and a gyroscopic platform to prevent tilting of the scanner. Usually, the navigation system is a differential global positioning system (DGPS) to determine the flight coordinates. An inertial navigation system (INS) is used to determine changes in aircraft attitude and to improve the position determination.
116
3.3 Nonphotographic Imaging
Table 3.3-2: Specifications of the thermography scanner AGEMA 900 from AGEMA Infrared Systems, Sweden, (from documentation of the Engineering and Physical Sciences Research Council (EPSRC), Great Britain, and the University of Magdeburg) spectral sensitivity filming speed number of pixels FOV
IFOV
temperature range thermal sensitivity
8 - 12 µm 15 Hz 136 × 272 20° × 10° 10° × 5.0° 5.0° × 2.5° 2.5° × 1.25° 1.28 mrad 0.64 mrad 0.32 mrad 0.16 mrad -30 °C to 1500 °C (2 000 °C with filter) 0.08 °C at +30 °C
Table 3.3-3: Specifications of the thermography scanner Thermo Tracer TH 1101 from NEC San-ei, Japan, (from documentation of the ebs GmbH, NEC San-ei distributor in Germany) spectral sensitivity number of pixels FOV IFOV temperature range thermal sensitivity
8 - 13 µm 344 × 207 30° × 27° 1.5 mrad -50 °C to 2000 °C 0.1 °C
117
Environmental Geology, 3 Remote Sensing Table 3.3-4: Specification of the Landsat sensors, ALBERTZ (2001)
in operation altitude repeat cycle [d] swath width pixel size spectral channels
Landsat 4 & 5 (1-3) Multispectral Scanning System (MSS)
Landsat 4 & 5 Thematic Mapper (TM)
Landsat 7 Enhanced Thematic Mapper (ETM+)
since 1972 705 km (915 km) 16 (18) 185 km 79 × 79 m² 1 (4) 0.50 - 0.60 µm 2 (5) 0.60 - 0.70 µm 3 (6) 0.70 - 0.80 µm 4 (7) 0.80 - 1.10 µm
since 1982 705 km 16 185 km 30 × 30 m² 0.45 - 0.52 µm 0.52 - 0.60 µm 0.63 - 0.69 µm 0.76 - 0.90 µm 1.55 - 1.73 µm 2.08 - 2.35 µm 10.4 - 12.5 µm 120 × 120 m²
since 1999 705 km 16 185 km 30 × 30 m² 0.45 – 0.52 µm 0.52 - 0.60 µm 0.63 - 0.69 µm 0.76 - 0.90 µm 1.55 - 1.73 µm 2.08 - 2.35 µm 10.4 - 12.5 µm 60 × 60 m² 0.52 – 0.90 µm 15 × 15 m²
thermal channel
1 2 3 4 5 7 6
panchromatic channel
1 2 3 4 5 7 6 8
Table 3.3-5: Specification of the SPOT-Sensors, ALBERTZ (2001)
in operation altitude repeat cycle [d] pixel size image size spectral channels
SPOT HRV (XS-mode) multispectral
SPOT HRV (P-mode) panchromatic
since 1986 832 km 26 * 20 × 20 m² 60 × 60 km² 1 0.50 - 0.59 µm 2 0.61 - 0.69 µm 3 0.79 - 0.89 µm
since 1986 832 km 26 * 10 ×10 m² 60 × 60 km² 0.51 - 0.73 µm
SPOT 4 (launched 1998) is equipped with an additional spectral channel in short-wave infrared (1.58 – 1.75 µm) with a pixel size 20 × 20 m². * Shorter repeat cycles are possible for certain areas by changing the look angle within +27° and -27°.
118
3.3 Nonphotographic Imaging
Table 3.3-6: Specifications of the Indian satellites IRS-IC and IRS-ID, ALBERTZ (2001)
launched altitude repeat cycle [d] resolution swath width spectral channels
WiFS
LISS-III
PAN
1996/1997 817 km
1996/1997 817 km
5
24
1996/1997 817 km 24 (5 days with a look angle of up to 26°) 5.8 m 70 km 0.50 - 0.75 µm
188 m 810 km 0.62 - 0.68 µm 0.77 - 0.86 µm 1.55 - 1.75 µm
23 m 142 km 0.52 - 0.59 µm 0.62 - 0.68 µm 0.77 - 0.86 µm 1.55 - 1.70 µm resolution 70 m
Table 3.3-7: Specifications of the OPS sensor system on Japan’s Earth Resources Satellite-1 (JERS-1) Spatial resolution
18.3 u 24.2 m
Detector
CCD linear array
Spectral range 0.52 0.63 0.76 0.76 1.60 2.01 2.13 2.27 -
0.60 µm 0.69 µm 0.86 µm 0.86 µm * 1.71 µm 2.12 µm 2.25 µm 2.40 µm
Swath width
75 km
* Forward viewing for stereo imaging. Besides the optical sensor (OPS), an SAR system is operated on the same satellite.
119
Environmental Geology, 3 Remote Sensing
Table 3.3-8: Specifications of the current high-resolution satellite sensors, ALBERTZ (2001) completed IKONOS 2 company launched altitude orbit inclination repeat cycle [d] minimal repeat cycle [d] resolution (panchromatic) band width resolution (multisp.) swath width spectral channels
Space Imaging 24.09.1999 680 km 98° 14 1-3 0.82 m 0.45 - 0.90 µm 4m 11 km 0.45 - 0.52 µm 0.52 - 0.60 µm 0.63 - 0.69 µm 0.76 - 0.90 µm
Quick Bird 2
OrbView 3
Earth Watch Inc. 18.10.2001 600 km 66° 20 1.5 - 2.5 0.82 m 0.45 - 0.90 µm 3.28 m 22 km 0.45 - 0.52 µm 0.52 - 0.60 µm 0.63 - 0.69 µm 0.76 - 0.89 µm
Orbital Image Corp. 26.06.2003 470 km 97° 16 1-3 1m 0.45 - 0.90 µm 4m 8 km 0.45 - 0.52 µm 0.52 - 0.60 µm 0.62 - 0.70 µm 0.76 - 0.90 µm
Table 3.3-9: Specifications of the Advanced Spaceborne Thermal Emission and Reflection (ASTER) sensor on the Earth Observation Satellite (EOS-AM-1), GUPTA (2003) Spectral band VNIR green red NIR NIR SWIR
TIR
1 2 3 3B 4 5 6 7 8 9 10 11 12 13 14 Swath width:
Spectral range
Scanner type
Spatial resolution
Signal quantization
0.52 - 0.60 µm 0.63 - 0.69 µm 5 000 cells 15 m 8-bit 0.76 - 0.86 µm 0.76 - 0.86 µm 1.60 - 1.70 µm 2.145 - 2.185 µm 2.225 - 2.245 µm 2 048 cells 30 m 8-bit 2.235 - 2.285 µm 2.295 - 2.365 µm 2.360 - 2.430 µm 8.13 - 8.48 µm 8.48 - 8.83 µm OM 90 m 12-bit 8.90 - 9.25 µm 10.25 - 10.95 µm 10.95 - 11.65 µm 60 km; band 3B is backward-looking for stereo viewing
120
3.3 Nonphotographic Imaging
Table 3.3-10: Satellite radar system specifications, ALBERTZ (2001)
company launched altitude orbit inclination sensor system mode resolution (panchromatic) swath width repeat cycle [d] resolution band depression angle polarization
ERS
RADARSAT-1
ENVISAT
ESA (European Space Agency) 1995 (ERS-2) 780 km 98.5° AMI (Active Microwave Instrument) image mode
CSA/CCRS (Canadian Space Agency) 1995 798 km 98.6°
ESA (European Space Agency) 2002 800 km 98.5°
SAR
ASAR (Advanced SAR)
standard mode
image mode
0.82 m
0.82 m
1m
100 km 35 25 m C (5.3 GHz/5.6 cm) ~ 23°
100 km 24 28 m C (5.3 GHz/5.6 cm) 20 - 49°
vertical/vertical
horizontal/horizontal
56 - 120 km 35 30 m C (5.33 GHz/5.6 cm) 15 - 45° vertical/vertical or horizontal/horizontal
Table 3.3-11: Specifications of the Shuttle Radar Topographic Mission (SRTM) dedicated to SAR interferometry for high-accuracy and high-resolution DEM, GUPTA (2003) Shuttle launched duration altitude
11 February 2000 11 days ~ 275 km orbit
SAR sensors wavelength frequency look angle swath width
C-band 5.6 cm 5.3 GHz 30 - 60° 225 km
X-band 3.1 cm 9.6 GHz 50 - 55° 50 km
121
Environmental Geology, 3 Remote Sensing
Table 3.3-12: Specifications of the High-Resolution Stereo Camera HRSC of the DLR German Aerospace Center (from DLR documents) HRSC-A focal length FOV
175 mm 38º × 12º 9 number of CCD lines (4 colors) pixel per line 5272 pixel size 7 µm radiometric resolution 8 bit max. scan frequency 450 lines s-1 spectral resolution and viewing axis (model AX)
HRSC-AX
HRSC-AXW
151 mm 47 mm 41º × 29º 30º × 79º 9 5 (4 colors) (2 colors) 12 172 12 172 6.5 µm 6.5 µm 12 bit 12 bit 1640 lines s-1 1640 lines s-1 blue: 440 - 510 nm -4.6° green: 520 - 590 nm -2.3° red: 620 - 680 nm 2.3° NIR: 780 - 850 nm 4.6° PAN: 520 - 760 nm 20.5°; -20.5°; 12°; -12°; 0°
Table 3.3-13: Specifications of the Digital Modular Camera of ZI Imaging, Germany (from documents of the manufacturer) FOV cross u along panchromatic: resolution, lens system multispectral: channels, resolution, lens system frame rate radiometric resolution
74° u 44° 13 500 × 8000 pixel, 4 × f = 120 mm 4 × RDG & NIR, 3000 × 2000 pixels, 4 × f = 25 mm 2 s/image 12 bit
Table 3.3-14: Spectral recording channels and band widths of the Daedalus AADS-1268 airborne scanner (after DLR documents) Channel number Spectral range 1 2 3 4 5 6 7 8 9 10 11
visible light
near infrared (I) near infrared (II) thermal infrared
Channel range in µm 0.420 - 0.450 0.450 - 0.520 0.520 - 0.600 0.605 - 0.625 0.630 - 0.690 0.695 - 0.750 0.760 - 0.900 0.910 - 1.050 1.55 - 1.75 2.08 - 2.35 8.5 - 13.0
122
3.3 Nonphotographic Imaging
Table 3.3-15: Selected specifications of the CASI scanner (from documentation of WIB GmbH, Berlin) angular field of view (FOV) number of detectors per scan line IFOV spectral range total number of spectral recording channels available maxium number of channels that can be simultaneously used
35.4q 512 1.21 mrad 0.43 – 0.87 µm 288 with 1.8 nm bandwidth 15
Table 3.3-16: Selected specifications of the HyMap hyperspectral scanner (from documentation of Integrated Spectronics, Australia) Bandwidth Average spectral across Module Spectral range sampling module interval VIS NIR SWIR1 SWIR2
0.45 - 0.89 µm 0.89 - 1.35 µm 1.40 - 1.80 µm 1.95 - 2.48 µm
15 - 16 nm 15 - 16 nm 15 - 16 nm 18 - 20 nm
15 nm 15 nm 13 nm 17 nm
IFOV
Signal-tonoise ratio
2.5 mrad along-track 2.0 mrad across-track
! 500 : 1
Table 3.3-17: Selected specifications of the Ekwan hyperspectral imager (from documentation of Ekwan Technology Corporation) IFOV number of spectral channels wavelength signal to noise ground resolution spectral bandwidth swath width
0.9 mrad 384 370 to 2495 nm > 400 : 1 (SWIR) 0.8 - 7 m 5.25 nm to 6.25 nm up to 2 km
123
Environmental Geology, 3 Remote Sensing
Table 3.3-18: Examples of current and future imaging spectrometers, SLONECKER & WILLIAMS (2001) Acronym
Name
Manufacturer
Bands
Spectral range [nm]
ASAS
Advanced Solid-State Array Spectrometer
NASA (Goddard)
62
400 - 1200
AIS-1
Airborne Imaging System 1
NASA (JPL)
128
800 - 1600 1200 - 2400
AISA
Airborne Imaging Spectrometer for Applications
SPECIM, Ltd.
286
400 - 1250
CIS
Chinese Imaging Spectrometer
Shanghai Institute of Technical Physics
91
-
GER Corp
79
400 - 12000
ERIM
256
2000 - 15000
Daedalus
102
433 - 12700
ASI
512
300 - 1050
NASA (JPL)
224
400 - 2500
NRL
210
400 - 2500
Integrated Spectronics
200
400 - 12000
Daedalus
50
530 - 14500
NASA
36
400 - 14400
TRW U.S. Navy U.S. Air Force
384 210 280
300 - 2500 400 - 2500 400 - 5000
Dais 7915 IRIS MIVIS VIMS-V AVIRIS HYDICE HYMAP MAS MODIS TRWIS III NEMO Warfighter1
Digital Airborne Imaging Spectrometer Infrared Imaging Spectroradiometer Multispectral Infrared and Visible Imaging Spectrometer Visible Infrared Mapping Spectrometer Advanced Visible and Infrared Imaging Spectrometer Hyperspectral Digital Imagery Collection Experiment Airborne Hyperspectral Scanner MODIS AIRBORNE SIMULATOR Moderate Resolution Infrared Spectrometer TRW Imaging Spectrometer Navy Earth Map Observer Warfighter
124
3.3 Nonphotographic Imaging
Table 3.3-19: Airborne laser scanner specifications, GIM International (2001) Name manufacturer scanning principle
ALTM 2033
Saab / TopEye
Optech Systems TopEye, Saab Corp. oscillating, rotating, oscillating, Z-shaped, Z-shaped, fibers mirror mirror, optical
TopoSys® II
ALS 40
TopoSys, Saab
LH Systems
fibers
oscillating, mirror
data rate
max. 33 000 Hz
max. 7000 Hz
max. 83 000 Hz
max. 30 000 Hz
scan rate
max. 66 Hz
6.25 - 25 Hz
650 Hz
20 - 26 Hz
scan angle
max. ± 20°
max. ± 20°
max. ± 7°
max. ± 37.5°
max. height
6000 m
960 m
1600 m
6100 m
15 - 25 cm
< 10 cm
height accuracy
< 15 cm 8 cm (depending on (depends on DGPS accuracy) flight altitude)
3.3.5 Survey Practice GEBHARDT (1981) discusses factors influencing and disturbing a thermal survey, e.g., wind, precipitation and clouds. Wind blurs thermal anomalies and can cause temperature effects that make the data unusable. Precipitation lowers thermal contrasts for a considerable period of time. Cloud cover causes atmospheric counter-radiation. This produces artifacts and reduces the contrasts between the anomalies on the ground. Vegetation prevents thermal evaluation of the soil and rocks and contributes its own thermal signature. The time after sunset is best suited for distinguishing thermal anomalies on the Earth’s surface and buried objects (e.g., waste, “hot spots” inside tailing dams, gas emissions). Nighttime is preferred because the temperature is more constant than during the day ensuring more representative data and improved interpretations. Thermal contrast observed during the day is mostly due to different amount of exposure to solar radiation. For investigation of geologic thermal phenomena, e.g., hot spots, thermal springs and gas emissions, temperatures slightly below zero are advantageous since thermal contrasts are enhanced. The time just after noon are most suitable for thermal localization of springs, seepage localities, drainage systems, irrigation ditches, etc. This is due to the high thermal inertia of water (Fig. 3.3-8), which causes these area to have a lower temperature than the area around it at this time. Higher evaporation and moisture-saturated air on top of a moist site can cause thermal minima as well.
Environmental Geology, 3 Remote Sensing
125
The conditions necessary for a digital camera survey flight are similar to those for an aerial photographic survey flight (Chapter 3.1). The weather should be clear without clouds or haze. A sophisticated navigation system, e.g., DGPS, provides precise image-coordinates in the cm-range. This helps to avoid additional ground surveys to obtain the control point coordinates. Imaging spectrometer measurements require high intensity solar radiation during the flight to allow sufficient signal for all channels of the instrument and thus improve the signal-to-noise ratio. Fog, mist, haze, smog, precipitation and all kind of clouds have to be avoided. Acquisitions near solar noon are the ideal time of day if there are no disturbing meteorological or man-made factors. Vegetation cover (season) should be avoided if the investigation targets are soils, rocks or their contamination. A calibration site and several additional sites should be carefully chosen before the flight. These are measured on the ground using a field spectrometer during the flight. The sites should be spectrally homogeneous (given the spatial resolution of the sensor) in order to obtain calibration spectra for the radiometric correction. The site has to represent a single possibly bright material and should be larger than one pixel of the remote-sensing image so calibration pixels contain a spectrum for only this material. The quality of such a standard field measurement determines the interpretation quality of the remote sensing data. For simplification, the calibration sites should be near the airport or on the premises of the client. Additional measurement of different objects within the area of interest further improves the data processing and ensures the reliability of the data interpretation. Some manufacturers build hyperspectral scanner systems with spectral ranges from visible light to the middle thermal infrared. If the study involves a larger spectral range it has to be considered that thermal surveys have different requirements than a survey constrained to the 0.4 – 2.5 µm range. For high-resolution satellite sensors (e.g., IKONOS, QuickBird), up-todate images for selected areas can be acquired and obtained. Acceptable time slots and percentage of cloud cover must be given along with the data order. The time window should not be too short in order to increase the chances for good weather conditions when the satellite passes over the target area. The smaller the desired percentage of cloud cover given in the order, the higher the cost of the images. On the other hand, clouds and their shadows may obscure objects of interest.
3.3.6 Processing and Interpretation of Data Sophisticated computers with an appropriate memory, powerful graphic cards, and large storage capabilities are an essential for the processing and interpretation of remote-sensing data. Another prerequisite is the availability of an effective software package for the following tasks:
126
3.3 Nonphotographic Imaging
x image rectification and restoration (preprocessing), x image enhancement, x image analysis, x image classification, x data merging, GIS integration and map preparation. Examples of powerful and versatile remote-sensing software systems are: x ENVI, x ERDAS, x Earth Resource Mapping / ER Mapper, x PCI, and x Tetracorder. Exemplarily some features of the ENVI system are listed: x Support of numerous data import and output formats, x data preprocessing and radiometric calibration, x image registration and orthorectification, x interactive image enhancement, x panchromatic, multi- and hyperspectral, and radar data analysis, x spectral analysis and filtering, x change analysis, x mosaic and subset, x supervised and unsupervised classification, x integrated GIS features, x annotation and map composition, and x terrain analysis.
Preprocessing Raw images from nonphotographic scanners contain systematic and nonsystematic geometric distortions. Sources of these distortions include variations in altitude, attitude, and velocity of the sensor platform, panoramic distortions, effects due to Earth’s curvature and rotation, as well as relief displacement. In airborne applications, data from the DGPS and INS
Environmental Geology, 3 Remote Sensing
127
navigation system are used to eliminate distortions from variations in the altitude, attitude, and velocity of the aircraft. These correction procedures are not necessary for optical satellite data. Due to the small aperture and the high altitude of space-borne imaging systems the relief displacement can often be neglected. For geoscientific interpretation, the images from space-borne systems can be used directly as delivered by the provider. Rectification and restoration of radar images requires sophisticated processing. A digital elevation model (DEM) is necessary for correction of topographic displacements. Preprocessing can be done by the provider of the satellite imagery. Satellite images are delivered and indexed in terms of path and row numbers for easy referencing and ordering. Web-based access allow the search in large satellite data archives. The images can be ordered in different qualities (e.g., standard and precision) as photographs (e.g., positive or negative image, false-color images) or as digital data.
Processing Thermal scanner manufacturers provide specific and customized software systems for processing and interpretation of the data. Images from thermal scanners can be handeled with the same hardware and software used to record and store the data during the flight. This easy-to-use windows-based software permits the generation of most suitable thermal images using different processing techniques, such as enhancement of temperature patterns by stretching, plotting isotherms, determination of temperature along profiles and at point sites, and color coding of temperature intervals. The processing of the data from digital cameras is described here using the example of the HRSC camera. The first processing step is always radiometric correction to minimize or at least reduce noise and to enhance the desired signal. The external orientation of the image lines is then computed using the post-processed data from a special flight management system that recorded the exact position of the aircraft every few seconds during the flight. Using the same points derived from the different stereo channels and from overlapping image strips, the angular offset between the Inertial Navigation System (INS) and the camera axis can be calculated and the orientation of the adjacent image strips along the flight line can be corrected. Calculations of a digital elevation model can now be completed from the set of object points. The terrain model is needed to generate orthoimages correcting for topographic displacements. The orthoimages are then combined to orthoimage-mosaics. The final step is color processing to generate color composites (color, CIR) of the ortho images or ortho image mosaics. All data processing is fully automated. The processing of data from imaging spectrometers always starts with a radiometric correction using laboratory standards. This step aims to calibrate
128
3.3 Nonphotographic Imaging
the electric signal of the spectrometer with respect to the radiance of the incident radiation. When the sensor has been calibrated, an atmospheric and other radiometric correction are performed. An atmospheric correction is advised because of the sensor’s sensitivity to changing atmospheric conditions, i.e. gases, water vapor, aerosols and other substances in the atmosphere. This correction is vital to suppress or eliminate this type of noise and to improve the signal-to-noise ratio. There are different models of atmospheric scattering and absorption available for this step. Further radiometric correction is necessary to match the signals measured by the imaging spectrometer to the spectrally acquired materials on the ground. After the radiometric correction, the absolute value of the radiance is converted into percentage reflectance of the incident radiation. The next step is a geometric correction using the orientation data from INS and DGPS. The quality of the geometric correction can be improved by using the data from a second sensor with a higher ground resolution, like a digital camera. Interpretation Interpretations of nonphotographic imaging data and the use of the results for site investigations depend on the spectral range of the sensors used for imaging. The solar reflection (SOR) region from 0.3 to ~3 µm seems the best investigated part of the electromagnetic (EM) spectrum. Such data are easy to interpret in terms of directly observable objects and physical phenomena. All types of sensors used on aerial and space-borne platforms acquire data in this spectral range. But, the images from these systems differ in spatial and radiometric resolution. The use of multi-spectral data from the SOR region facilitates geological and environmental investigations and monitoring, such as topography and drainage patterns, identification of rock and soil types, moisture content, faults, and vegetation. Given the thermal conditions of the Earth, EM radiation with wavelengths from 3 to 35 µm is emitted from the surface. The amount of energy an object on the ground radiates depends on its surface temperature and emissivity. Emissivity is a materials property. Previous investigations of waste disposal sites have shown that the thermal conditions depend on the time of day, time of year, and the weather prior to data collection; the usability and reproducibility of aerial photographs and thermal images are also related to these parameters. Qualifiers, such as warm or cold, can be used only together with a specific time of day because diurnal temperature variations differ for different materials (Fig. 3.3-8). Consequently, generalized statements regarding the thermal properties of materials in the Schöneiche landfill (e.g., EHRENBERG, 1991) may be misleading.
Environmental Geology, 3 Remote Sensing
129
Fig. 3.3-8: Surface temperatures of water and grass during a diurnal cycle, LOWE (1969)
To maximize the usefulness of thermal remote-sensing data, surface temperatures of rocks, soils, vegetation, and waste should be determined using an array of ground sensors at the time of remote data take. Emissivities of samples may also be measured in the laboratory. One objective of thermal imagery is the detection of heat sources within mine waste heaps (“hot spots”), for example produced by dumping of hot ashes and highly inflammable material (Figs. 3.3-11, -12, and -13). If the amount of smoldering material is large enough and the internal temperature is high enough, such phenomena can be detected even when covered up to several meters. Fermentation of organic matter in a landfill produces heat that is transported to the surface by methane and other gases. Ducts are installed to remove these gases under controlled conditions (Fig. 3.3-16). Uncontrolled emissions are also possible and can be detected using thermal imagery. Another application of thermography is the detection of buried waste. Because the heat capacity of waste is usually different than surrounding natural material, the waste is revealed by thermal observations (Figs. 3.3-17 and 3.3-18). It is well known that the thermal inertia and heat capacity of water is different from those of soils and rocks. This information is used to detect natural and artificial features containing water. Such features can be springs and soil moisture seeps in areas with high groundwater levels or in areas with non-cohesive soil (Figs. 3.3-19 and 3.3-20), filled-in natural and artificial drainage systems (Fig. 3.3-21), seepage water, and fractures and faults in karst regions. The combination of a DEM with thermal data can reveal indications of night time cold air fluxes and natural ventilation. A sufficient slope of the landfill and a sparse vegetation cover are required conditions for such processes to happen. Cold air flowing from the top of the landfill is indicated in night images by "cold" anomalies along the slope. Such cold air flows can
130
3.3 Nonphotographic Imaging
possibly carry harmful substances that would then accumulate in depressions around the landfill. Data from digital cameras can be interpreted independently or in combination with other remote-sensing data. Photogrammetric products (orthophotos, DEMs, orthophoto mosaics, topographic maps, etc., Chapter 3.2) or topographic base maps for field work and the presentation of results can be produced from digital camera images. The spectral channels of the camera permit a spectral composition similar to color and CIR photographs and the integration of the images with other data sets, e.g., with a DEM (Fig. 3.3-24). There are a range of applications like regional land-use mapping and mapping of biotopes to site investigations, for example, to evaluate the vitality of vegetation, assess surface water conditions, or search for seepage water, springs and moisture anomalies (Chapter 3.1). Yet, the spatial resolution of an analog aerial photograph, depending on the film granularity, is still better than the spatial resolution of most digital cameras. Hyperspectral images particularly allow for identification of certain subpixel features. Most hyperspectral pixels have a mixed spectral signature, originating from different objects or materials. If this mixture can be detected in the pixel spectrum due to the presence of different spectral signatures, the spectral signatures can be assigned to different objects and their areal coverage (spectral unmixing). In order to recognize and distinguish materials in the imaging spectrometer data, local measurements need to be made with a field spectrometer. This is necessary to determine whether the contaminated materials can be spectrally discriminated from the surrounding uncontaminated materials and thus mapped from the image data. Minerals and mining waste exposed in open-pit mines, at mill sites, mine dumps, tailing and spoil heaps, etc. have to be identified and mapped to characterize large regional mining districts. LIVO (1994), JANSEN (1994) and PETERS & PHOEBE (2000) developed a method for this purpose using the AVIRIS system achieving a spatial resolution of 20 m in mining areas of North America. KING et al. (2000) discussed the impact of water and wind on mining districts. Sulfides constitute a significant part of the mineral paragenesis of such districts, as well as many other ore minerals and resulting weathering products. Pyrite and alunite generate sulfuric acid in contact with rain water seeping into the groundwater. Such acidic sites in mining districts can be detected by mapping the resulting mineral alunogen. Using specific software, the spectra of the image pixel are compared with standard spectra of the digital spectral libraries as provided by the USGS (U.S. Geological Survey). The spectral range for discriminating minerals is often focused on 2200 - 2300 nm. Visual examination and modern algorithms for spectral identification can be used by an experienced specialist. Such methods not only allow to identify pixels with the spectrum for a single material, but also pixels
Environmental Geology, 3 Remote Sensing
131
showing the spectra of two minerals, for example dolomite and calcite, kaolinite and hematite, jarosite and muscovite and other materials.
(a)
(b)
Fig. 3.3-9: Radiance spectra of MARK V laboratory spectrometer (a) and HyMap spectra (b) of reference areas on BGR premises
132
3.3 Nonphotographic Imaging
Tests with hyperspectral images in different mining areas with similar contaminated materials indicate that a simple transfer of the method is not possible. The different climatic, weathering, geomorpological and vegetation conditions in other regions and the mineral mixtures, man-made materials and vegetation cover typical for those regions point at the importance to collect spectroscopic data in the field and laboratory, hence regional spectral libraries as a basis for reliable calibration and interpretation. The HyMap airborne imaging spectral system has been used (Figs. 3.3.-9 and 3.3-22). to detect hydrocarbons (HÖRIG et al. 2001) Hydrocarbons and any materials containing them are characterized by absorption maxima at wavelengths of 1730 and 2310 nm (CLOUTIS, 1989). The radiance spectra recorded by a Mark V laboratory spectrometer (Fig. 3.3-9a) had the same two absorption peaks as shown in the spectra of the airborne system recorded simultaneously (Fig. 3.3-9b). The 1730 nm feature is more pronounced. Better results for identifying the hydrocarbons were obtained using this absorption peak than with the 2310 nm peak. The interpretation uses at least three bands of the HyMap which cover the portion of the spectrum around these two peaks.
Fig. 3.3-10: Processed HyMap image showing materials containing hydrocarbon at the ground surface in Spandau, Berlin (BGR premises and oil-contaminated reference areas; A, artificial turf; T, race track; B, plastic roofs)
Environmental Geology, 3 Remote Sensing
133
A test area in Spandau, Berlin is shown in Figure 3.3-10. Most of the surface materials in the images appear grey because spectral characteristics are similar in this part of the spectrum. Only the areas containing hydrocarbons appear colored due to their significant absorption features within the same narrow spectral range. Consequently, oil-contaminated soil reference areas (southern part of the picture, near BGR), plastic roofs of buildings (B), artificial turf (A), and athletics race tracks (T) appear pinkish. The intensity of the color of the area depends probably on the oil content. ELLIS et al. (2001) used airborne hyperspectral imaging data from Probe-1, the preceding model of HyMap to detect several onshore oil seeps. Earlier mapped oil seeps were confirmed by these measurements and new sites were discovered. The "depth" of the absorption minimum in both the field and airborne spectra at 2310 nm correlates well with the amount of bitumen (or tar and asphalt). The main objective of further investigation is the compilation of a spectral library of hydrocarbon maxima and minima, including spectra with different amounts of oil, tar, vegetation, soil and rock. Using this approach, it is possible to map contaminations with oil, gasoline/petrol, kerosene, diesel, etc. even for large areas. Natural oil seeps, natural hydrocarbon-bearing sediments, e.g., oil shale or sandstone, and artificial plastic hydrocarbon materials, may be detected. No chemical laboratory method is able to provide such a spatial database as quickly, reliably and as inexpensively as this remote sensing method. Radar images from various aerial and space-borne SAR sensors provide information about changes in the terrain slope on a decameter scale, of changes in surface roughness on a centimeter scale, and about electrical properties of the materials on or near the Earth’s surface. Information about the soil moisture and mineral content of soil and rock can be derived from the dielectric properties. A large number of studies have shown that interferometric SAR (InSAR) provides very good data for detecting vertical ground displacement on a centimeter scale (Fig. 3.3-25). An overview of current InSAR applications for land subsidence detection is available at various websites, e.g., http://www.npagroup.com and http://www.atlsci.com.
134
3.3 Nonphotographic Imaging
3.3.7 Quality Assurance To obtain high-quality results, an approriate remote-sensing system has to be selected given the objectives of the investigation. This decision depends on the spatial and radiometric resolution, the spectral range needed and the expected cost. This is the most important factor that determines the quality of the data and data products. In addition, the quality is determined by the following conditions: x technological factors, x conditions during data acquisition, and x processing and interpretation. Technological factors: 1. Calibration of the nonphotographic remote-sensing systems must be done before and during data acquisition following the instructions of the manufacturer. 2. INS (Inertial Navigation System) must be used for accurate determination of changes in aircraft attitude around all three axes and for correction the GPS data. 3. Data with full Image Motion Compensation (IMC) or Forward Motion Compensation (FMC) should be preferred if airborne systems are used. 4. High precision DGPS with a ground reference station closer than 50 km from the project area is strongly advised. 5. For laser scanning flights, DGPS and INS are essential in order to estimate the precise aircraft position at any point. A video camera should be used to record the path of the aircraft. A synchronization system is necessary to correlate the data with other data acquisition systems. Conditions during data-sampling: 1. Weather: x Thermal scanner: no wind, clouds, mist, or precipitation, as well as complete clear conditions. x Digital camera: sunny, without clouds, mist or precipitation. x Imaging spectrometer: sunshine with no fog, mist, haze, smog, clouds, or precipitation, etc. to reduce the intensity of the sun’s radiation.
Environmental Geology, 3 Remote Sensing
135
2. Time of data acquisition: x Thermography: season with only little or no growing vegetation, during the night or in the early morning before sunrise following a sunny day. x Digital camera: depends on the investigation objectives, e.g., season with or without growing vegetation. x Imaging spectrometer: season with only little or no growing vegetation, the hours around high noon are preferable. Processing and interpretation: x Experience and references: The company or institution conducting the survey should provide references in carrying out remote-sensing surveys using state-of-the-art hardware and software. x Integration of other data: Including other remote-sensing and non remotesensing information increases the reliability and value of the results. x Ground truth: Thermograpic and spectroscopic measurements are made on the ground at the time of the survey flight. If airborne digital camera data is acquired during the flight, additional ground truth data may be required for the interpretation.
136
3.3 Nonphotographic Imaging
3.4.8 Personnel, Equipment, Time Needed Personnel data acquisition with thermography scanner processing data from a thermography scanner interpretation of thermography data data acquisition with a digital camera processing data from a digital camera (HRSC) interpretation of data from a digital camera (HRSC)
special survey aircraft with a 2 - 3 technicians flight management system and or specialists a thermography scanner
Time needed 1 day
1 specialist
PC, plotter, software
1 - 3 days
1 specialist
PC, plotter, software
3 - 8 days
special survey aircraft with a 2 - 3 technicians flight management system and or specialists a digital camera system 1 specialist
1 specialist
groundwork for imaging 1 - 2 technicians spectrometry or specialists data acquisition with an imaging spectrometer
Equipment
2 - 3 technicians or /specialists
1 day
several PCs, plotter, software
1 - 3 weeks
PC, plotter, software
1 day to some weeks (depends on the task)
1 field spectrometer
1 - 3 days
special survey aircraft with a flight management system, an imaging spectrometer and a digital camera
1 day
measurements at the 1 - 2 technicians calibration site for 1 field spectrometer or specialist imaging spectrometry processing data of 1 specialist PC, plotter, software imaging spectrometer interpretation of imaging spectrometer data 1 specialist PC, plotter, software (mapping of hydrocarbons) special survey aircraft with a data acquisition with an 2 - 3 technicians flight management system and airborne laser scanner or specialists laser scanning equipment preparation of a digital elevation model by laser 1 specialist PC, plotter, software scanning (area 100 km²)
1 day 1 - 3 weeks
1 - 3 weeks
1 day
3 - 5 weeks
Environmental Geology, 3 Remote Sensing
137
3.3.9 Examples Waste disposal sites on a SPOT image
Fig. 3.3-11: Schöneiche and Schöneicher Plan waste disposal sites (inside dashed lines), south of Mittenwalde near Berlin, shown on a SPOT image taken on August 25, 1990 in panchromatic mode. Courtesy of FKP Ingenieurgesellschaft für Fernerkundung, Photogrammetrie, Kartographie und Vermessung mbH (Consulting Engineers in Remote Sensing, Photogrammetry, Cartography, and Surveying), Berlin
138
3.3 Nonphotographic Imaging
Preparation of a map of flood prone areas Landsat 7 data from May 3, 2000, were used for mapping flood-prone areas in the Dan Khun Thot region of Thailand in order to classify preselected areas for a potential waste disposal site. Image quality was good with cloud coverage of about 20 - 25 %. The satellite remote sensing systems used scanners for 7 multi-spectral bands. Ground resolution was between 15 × 15 m and 60 × 60 m. The data were corrected for systematic and radiometric errors. Purpose of the processing was to provide the interpreter with enhanced imagery for mapping flooded areas. A combined approach of unsupervised and supervised classification was chosen to classify selected land cover types in the area. This classification divided the pixels into statistically defined classes for certain types of relevant land cover: standing water, swamps, elevated areas, plantations, soil salinization areas and settlements. A final map at a scale of 1 : 50 000 was derived (Fig. 3.3-12).
Environmental Geology, 3 Remote Sensing
139
Fig. 3.3-12: Dan Khun Thot region of the Nakhon Ratchasima province of Thailand, map of flood-prone areas prepared using Landsat 7 data from May 3, 2000, showing the locations of proposed waste disposal sites
140
3.3 Nonphotographic Imaging
Localizing heat sources inside a mine spoil heap Mining activities commonly result in changes to the landscape. Investigations are necessary when risks are identified that may threaten people, infrastructure and natural resources. The detailed black-and-white aerial photograph in Figure 3.3-13 depicts parts of an abandoned open-pit mine in Thuringia, Germany. Mine dumps, mill tailings, tailing ponds, and abandoned mining equipment are visible. The mill tailings contain smoldering pyrite and carbonrich schist. The smoldering is sustained by oxygen seeping through fissures and crevasses in the heap. This may lead to contamination of the groundwater if rainwater percolates through the heap (Fig. 3.3-15).
Fig. 3.3-13: Aerial photo taken April 24, 1995, showing a mining landscape in Thuringia, Germany. The dominant landscape structures include part of an abandoned open-pit mine (being filled with tailings), waste heaps, and an abandoned smelting plant (Source: Landesvermessungsamt Thüringen).
Environmental Geology, 3 Remote Sensing
141
The thermal scanner image in Figure 3.3-14, taken at night, displays the area visible in the black-and-white aerial photograph of Figure 3.3-13. Areas of high temperature in the heap are indicated in the thermal image although the smoldering material is several meters deep inside the heap. The range of apparent surface temperatures (in °C) depicted in the image is indicated by the scale in the lower left corner. The yellow color denotes temperatures between 25 and 35 °C, the maximum. Monitoring of waste heaps with thermal systems provides information on the temporal and spatial extent of the oxidation zone. Rainwater percolating through smoldering pyrite and carbon-rich schist forms sulfuric acid, which will then contaminate the groundwater. These smoldering zones must be identified for effective remediation. The situation and the method are explained in Figure 3.3-15.
Fig. 3.3-14: Thermal image taken at 03:45 a.m. on May 23, 1994, of the abandoned mining site shown in Fig. 3.3-13. Temperature anomalies (bright) indicate areas of smoldering pyrite and carbon-rich schist measured through several meters of cover. These zones of buried smoldering material are marked by significant temperature anomalies at the ground surface. (Image: F. KÜHN, BGR).
142
3.3 Nonphotographic Imaging
Fig. 3.3-15: Smoldering pyrite and carbon-rich schist generate heat within a mine waste heap. The smoldering is sustained by an influx of oxygen through fissures and crevasses in the heap. Sulfuric acid is produced, which may lead to groundwater contamination. The buried smoldering zones are easily detectable by thermal remote sensing (Fig. 3.3-14).
Gas emissions Anaerobic conditions inside a landfill favor the production of methane and other gases by fermentation. In the northern part of the Schöneiche landfill (Fig. 3.3-11), a network of ducts is used to collect and disperse the gas. Where the gas ducts emerge, they are observable as “hot spots” in thermal images taken at night, when the landfill surface cools off rather rapidly. During the daytime, solar radiation heats the surface of the landfill and the ducts appear colder than the surrounding areas. These gas ducts are the only source of information about temperature anomalies and the decay processes within the landfill (Fig. 3.3-16).
Environmental Geology, 3 Remote Sensing
143
Fig. 3.3-16: A thermal image made at 5:15 CET on May 13, 1993, shows four emerging ducts (SP01 – 04) used for the dispersal of biogases. AGEMA 900 software for image processing and evaluation permits calculation of the "radiation temperature" at the gas ducts, which fluctuates between 17.8 and 19.6 qC. The temperature increase of 1.8 qC between SP03 and SP04 indicates a temperature increase within the landfill. (1) Slope, (2) covered waste, and (3) roads. (Image: F. BÖKER and F. KÜHN, BGR)
Abandoned landfills and areas of suspected contamination The detection of soil and groundwater contamination raises questions concerning the temporal and spatial dispersal of contaminants. Because remediation can be expensive, and the polluter is legally required to cover the cost, the landfill operator will want to establish the actual sources of the contamination to avoid charges of clandestine waste disposal that often occur at large sites. In 1992, the Working Group for Remediation and Management of Waste Disposal, Berlin, (Arbeitsgemeinschaft “Sanierungs- und Betriebskonzept/Abfallentsorgung Berlin”) conducted a survey of abandoned waste disposal sites and other areas of suspected contamination near the Schöneiche and Schöneicher Plan landfills (Fig. 3.3-11). A number of current and abandoned waste disposal sites, as well as sewage and waste water irrigation systems, were observed and mapped. The following example demonstrates how aerial photographs and thermal images are applied to detect and characterize an abandoned landfill. An elevated area, typical for landfills, can be identified between the Schöneiche and Schöneicher Plan sites in the center of the stereo-pair photographs in Figure 3.3-17. The area is near the southeast rim of a water-filled pit (black). It shows patterns typical of cropland. Additional information was provided by
144
3.3 Nonphotographic Imaging
thermal data (Fig. 3.3-18): The thermal image reveal temperature anomaly patterns oriented more or less N-S in this area, most likely indicating nearsurface changes of material (dark blue and red patches). NW–SE-oriented yellowish strips are due to the grain crop. The image helped define areas for taking samples to check the character of materials dumped in the area.
Fig. 3.3-17: CIR aerial photograph taken on July 3, 1993, (stereo-pair) showing an abandoned landfill (area in the center just below the body of water) between the Schöneiche and Schöneicher Plan landfills (Photo taken by WIB GmbH Berlin for BGR)
Fig. 3.3-18: Thermal image taken at 05:27 CET on May 13, 1993, showing the landfill (within dashed line) in Fig. 3.3-17 and temperature anomalies resulting from a partial vegetation cover (yellow and orange stripes) and from soil (dark blue stripes trending roughly N–S). Image taken by F. BÖKER and F. KÜHN, BGR
Environmental Geology, 3 Remote Sensing
145
Springs and soil moisture anomalies As discussed previously, moist areas and springs are easily detected by remote-sensing techniques. Moist areas commonly develop at the surface due to the presence of perched groundwater or to a shallow water table. Geochemical sampling of moist areas may identify the sources of the moisture by their geochemical signatures. The historical aerial photograph from 1974 in Figure 3.3-19 shows an area about 100 m northwest of the Schöneiche landfill. A system of ditches can be seen that are partially filled with silt (D), indicating that this area must be permanent wetland. This assumption is supported by thermal images of the site (Fig. 3.3-20). Signs of elevated water content can be seen in the thermal images. A "cold" (dark blue) zone indicates high moisture content in the topsoil. Electrical conductivities around 1200 µS cm-1, which is about four times the normal value, were measured in groundwater samples from this zone. This suggests that mineralized groundwater contains leachate from the landfill about 200 meters north of the waste site. Dark patches (P) in the aerial photograph (Fig. 3.3-19) also suggest surface inhomogeneity. A related ground check confirmed this feature to be an abandoned landfill. The extent to which the abandoned landfill contributes to the elevated groundwater mineralization is yet to be investigated.
Fig. 3.3-19: Aerial photograph taken on April 20, 1974, of an area just north of the Schöneiche landfill. The dark gray areas indicate higher surface moisture content. The water has been drained via partially silted-up ditches (D). The area of Fig. 3.3-20 is marked by the thin dashed line and the margin of the landfill is indicated by the thick dashed line. Relatively dark patches (P) indicate an abandoned landfill. Photo from Bundesarchiv (Federal Archives of Germany)
146
3.3 Nonphotographic Imaging
Fig. 3.3-20: Thermal image taken at 05:22 CET on May 13, 1993, of the area marked in Fig. 3.3-19. Dark areas correspond to low temperatures, indicating a higher soil moisture content (Image: F. BÖKER and F. KÜHN, BGR)
Natural and artificial drainage systems A flat lowland adjacent to a waste disposal site is shown in a black-and-white aerial photograph in Figure 3.3-21 (top). Geologically speaking, this area consists of humic sand, mud, and bog lime deposits. The aerial photograph was taken at the beginning of intensive plant growth and relatively high soil moisture content (mid-May). Weak features that may indicate an artificial drainage system (arrow) can be seen. The thermal image (MIR-II) in Figure 3.3-21 (bottom), taken under similar conditions, depicts the surface drainage more clearly because the drained ground has a higher temperature than the surrounding soil with a higher moisture content. In this case, the intensity of temperature anomalies are indicators of the efficiency of the drainage system. If there are signs of artificial or natural drainage or irrigation systems on or near a landfill, the site should be inspected to determine whether leachate from the landfill is drained by them.
Environmental Geology, 3 Remote Sensing
147
Fig. 3.3-21: Black-and-white aerial photograph taken May 15, 1992 (top), and a daytime thermal image, MIR-II (TIR), taken May 11, 1993 (bottom). The fishbone pattern of an artificial drainage system is weakly visible in the aerial photograph (arrow). This pattern is better emphasized in the thermal image. The soil is relatively dry and "warm" where the drainage system still functions, (aerial photo: Luftbildsammelstelle der Landesvermessung und Geobasisinformation Brandenburg; thermal image: F. BÖKER and F. KÜHN, BGR)
148
3.3 Nonphotographic Imaging
Identification and mapping of hydrocarbons and materials containing hydrocarbons A case study performed to determine the possibility to identify and to map hydrocarbons and materials containing hydrocarbons. For this purpose reference areas were prepared containing different materials (Table 3.3-20). The ability to distinguish materials containing hydrocarbons from other materials is based on the absorption peaks of hydrocarbons that occur within a narrow range of the short-wave infrared (SWIR) spectrum (see Section 3.3.6). Figure 3.3-22 shows the result of HyMap data analysis. In this part of the SWIR spectrum, non-hydrocarbons appear gray, e.g., uncontaminated sand and sewage sludge. Only the hydrocarbons appear colored, due to their significant absorption within this narrow portion of the spectrum. The reference areas containing hydrocarbons smaller than 4 m × 4 m are below the spatial resolution of the HyMap system and, thus, appear gray too. The black tarpaulin acts as a black body reflector. Table 3.3-20: Reference areas on HyMap images shown in Fig. 3.3-22 1 - roofing felt (4 m u 4 m)
10 - sand, uncontaminated (4 m u 4 m)
2 - sand mixed with CaCO3 (4 m u 4 m)
11 - sand, uncontaminated (6 m u 6 m)
3 - sand, slightly oil-contaminated 1) (4 m u 4 m)
12 - sewage sludge (4 m u 4 m)
13 - sewage sludge (4 m u 4 m) 4 - sand, highly oil-contaminated 2) (4 m u 4 m) 14 - sewage sludge (4 m u 4 m) 5 - sand, highly oil-contaminated 2) (2 m u 2 m) 16 - plastic tarpaulin, orange (6 m u 8 m) 6 - sand, highly oil-contaminated 2) (1 m u 1 m) 17 - plastic tarpaulin, black (4 m u 8 m) 7 - sand, uncontaminated, moist (4 m u 4 m) 21 - pile of transparent plastic sheet 8 - sand, uncontaminated (1 m u 1 m) 9 - sand, uncontaminated (2 m u 2 m)
1) 25 mL oil per 1 kg sand 2) 100 mL oil per 1 kg sand
Environmental Geology, 3 Remote Sensing
149
Fig. 3.3-22: Reference areas on processed and enlarged HyMap images (SWIR-1 bands 21/red, 26/green, 31/blue with linear stretching)
Identification and mapping of materials containing hydrocarbons by merging the data from two remote-sensing platforms (contribution by Tilmann Bucher) Hyperspectral data (HyMap) can be processed together with data from a digital camera (HRSC-A) to merge high spatial and high spectral information. The HyMap system provides data for the location of hydrocarbons. The HRSC-A camera provides precise geometrically corrected images with a spatial resolution finer than 1 m (Fig. 3.3-23) and a DEM for 3-D visualization (Fig. 3.3-24). Characteristic absorption features in the SWIR range can be used to map materials containing hydrocarbon (HÖRIG et al., 2001). This method was evaluated at the overburden heap of the Espenhain lignite mine located in the Central German Lignite Mining District in Eastern Germany. The HyMap and the HRSC-A data of this area have been recorded in August 1998, with a ground resolution of about 8 m for the HyMap data and 30 cm for the HRSCA data (Fig. 3.3-23). Figure 3.3-24 depicts the merge of the two data sets. A white triangle is clearly visible in the center of the lower picture, marking the hydrocarbonbearing area. One side of the triangle is more than 300 m long. The DEM shows the dimensions of the dump in this mining district and the elongated crests; a deposition product of the swinging wings of the mining excavator. The crests are within the hydrocarbon-free and in the hydrocarbon-bearing area, thus, the contamination with hydrocarbons occurred after the deposition of the crests.
150
3.3 Nonphotographic Imaging
Fig. 3.3-3-23: Overlay of high-resolution HRSC-A data (spatial resolution 30 cm, center) with HyMap data (spatial resolution 8 m) from an area near Espenhain, Germany. Clouds (dark) cover parts of the area
Fig. 3.3-24: An image produced for visualization by the merging of data from different sources. A high-resolution HRSC-A image (top) and a false color HyMap image (bottom) are draped over the HRSC-A DEM. The white areas (bottom) show strong indications of hydrocarbon absorption
Environmental Geology, 3 Remote Sensing
151
InSAR-based land subsidence map for Bangkok, Thailand Traditional leveling at benchmark stations are usually used to monitor land subsidence resulting from extraction of groundwater from the main aquifer in the Greater Bangkok area. Land subsidence maps derived from leveling show precise results for each benchmark station, but the areal distribution of subsidence is widely inaccurate. Furthermore, the leveling data were commonly collected over long periods of time depending on the size of the observation area and the number of benchmark stations. Thus, it is sometimes difficult to compare data from different stations with one another. An InSAR-based land subsidence map showing the subsidence between February 20 and October 23, 1996, (left) is compared in Figure 3.3-26 with the land subsidence map derived from conventional leveling in the Greater Bangkok area in 1995/1996. The InSAR-based land subsidence map is more accurate in terms of the two-dimensional representation of subsidence than the map made from benchmark leveling measurements. A combination of precise benchmark leveling and InSAR-based mapping provides the best results for the investigation of land subsidence. For details see KÜHN et al. (2004).
Fig. 3.3-25: (left) Interferogram calculated from ERS 1/2 data recorded on 21 and 22 December 1999 used for correction of the contribution of topography to the phase and (right) differential interferogram derived from ERS 1/2 data recorded on 20 February and 23 October 1996
152
3.3 Nonphotographic Imaging
Fig. 3.3-26: InSAR-based land subsidence map showing subsidence between February 20 and October 23, 1996, (left) compared with the land subsidence map derived from conventional leveling in 1995/1996 by BONTEBAL (2001) (right)
References and further reading ALBERTZ, J. (2001): Einführung in die Fernerkundung, Grundlagen der Interpretation von Luft- und Satellitenbildern. Wissenschaftliche Buchgesellschaft, Darmstadt. ANGER, C. D., BABEY, S. K. & ADAMSON, R. A. (1990): A new approach to Imaging Spectroscopy. Proceedings of SPIE, 72, 72-86. AZCUE, J.M., MURDOCH, A., ROSA, F., HALL, G. E. M., JACKSON, T. A. & REYNOLDSON, T. (1995): Trace elements in water, sediments, porewater, and biota polluted by tailings from an abandoned gold mine in British Columbia, Canada. J. Geochem. Explor., 52, 25-34. BABEY, S. K. & ANGER, C. D. (1989): A Compact Airborne Spectrographic Imager (CASI). IGARSS Proceedings, 2, 1028-1031. BERGER, B. R. (1986): The geological attributes of Au-Ag-base metal epithermal deposits. In: ERICKSON, R., comp., Characteristics of mineral-deposit occurrences, USGS Open File Report, 82-795, 119-126.
Environmental Geology, 3 Remote Sensing
153
BIANCI, R., CAVALLI, R. M., FIUMI, L., MARAINO, C. M. & PIGNATTI, S. (1997): Airborne remote sensing: Results of two years of imaging spectrometry for the study of environmental problems. In: Remote Sensing ‘96, Balkema, Rotterdam, 269-273. BOLDT, C. M. K. & SCHEIBNER, B. J. (1987): Remote sensing of mine waste. U.S. Bureau of Mines Information Circular 9152. BONTEBAL, M. (2001): Land Subsidence in Bangkok: An Overview of Changes in Land Subsidence over the last 25 Years. M. Sc. Thesis, Free University of Amsterdam, Netherlands. BRITTON, B. (2002): Uses of Satellite Imagery for Planning or Site Selection, EOM, 11, 2. BROWN, R. L. (1991): Cripple Creek then and now. Sundance Publications, Ltd., Denver. CHUCHIP, K. (1997): Satellite Date Analysis and Surface Modeling for Land Use and Land Cover Classification in Thailand. Berliner Geographische Studien, 46, Berlin. CLARK, R. N. (1999): Spectroscopy of Rocks and Minerals, and Principles of the Spectroscopy, In: RENCZ, A. N. (Ed.): Remote Sensing for the Earth Sciences, 1, 3-58. CLARK, R.N. (1999): Spectroscopy of Rocks and Minerals and Principles of Spectroscopy, In: A.N. RENCZ (Ed.). Manual of Remote Sensing, Chapter 1, Wiley and Sons, New York, 3-58. CLARK, R. N., GALLAGHER, A. J. & SWAYZE, G. A. (1990a): Material absorption band depth mapping of imaging spectrometer data using a complete band shape least-squares fit with library reference spectra. In: Proceedings of the Second Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) Workshop, JPL Publication 90-54, 176-186. CLARK, R. N., KING, T. V. V., AGER, C. & SWAYZE, G. A. (1995): Initial vegetation species and senescence/stress mapping in the San Luis Valley, Colorado using imaging spectrometer data. In: Summaries of the Fifth Annual JPL Airborne Earth Science Workshop, January 23.–26., R.O. Green, Ed., JPL Publication 95-1, 35-38. CLARK, R. N., KING, T. V. V., KLEJWA, M., SWAYZE, G. & VERGO, N. (1990b): High Spectral Resolution Reflectance Spectroscopy of Minerals. J. Geophys. Res., 95, 12653-12680. CLARK, R. N., SWAYZE, G. A., KOCH, C., GALLAGHER, A. & AGER, C. (1992): Mapping Vegetation Types with the Multiple Spectral Feature Mapping Algorithm in both Emission and Absorption. In: Summaries of the Third Annual JPL Airborne Geosciences Workshop, 1, AVIRIS Workshop, held in Pasadena, CA, on June 1.-5., 1992. JPL Publication 92-14, 60-62. CLARK, R. N., SWAYZE, G. A., GALLAGHER, A., KING, T. V. V. & CALVIN, W. M. (1993): The U. S. Geological Survey, Digital Spectral Library, 1, 0.2 to 3.0 µm, USGS, Open File Report 93-592, 1340, (Also published as a USGS Bulletin, 1300+pp, 1996). CLARK, R. N., SWAYZE, G. A., LIVO, K. E., KOKALY, R. F., SUTLEY, S. J., DALTON, J. B., MCDOURAL, R. R., & GENT, C. A. (2003): Imaging spectroscopy: Earth and planetary remote sensing with the USGS Tetracorder and expert systems, J. Geophys. Res., 108 (E12), 5131, doi: 10.1029/2002JE001847, December, 2003. http://speclab.cr.usgs.gov/PAPERS/tetracorder
154
3.3 Nonphotographic Imaging
CLOUTIS, E. A. (1989): Spectral Reflectance Properties of Hydrocarbons: Remote Sensing Implications. Science, 245, 165-168. COLWELL, R. N. (Ed.) (1983): Manual of Remote Sensing, 1, Theory, Instruments and Techniques, 2, Interpretation and Applications. American Society of Photogrammetry, Falls Curch. CONWAY, E. D. (1997): An Introduction to Satellite Image Interpretation. Johns Hopkins University Press. COCKS, T., JENSSEN, R., STEWART, A., WILSON, I. & SHIELDS, T. (1998): The Hymap™ Airborne Hyperspectral Sensor: The System, Calibration and Performance. In: SCHAEPMAN, M. et al. Eds. 1st EARSeL Workshop on Imaging Spectroscopy, 37-42. CRAWFORD, M. F. (1987): Preliminary Evaluation of Remote Sensing Data for Detection of Vegetation Stress Related to Hydrocarbon Microseepage: Mist Gas Field Oregon. Proc of 5th Thematic Conf on Remote Sensing for Exploration Geology, Environmental Research Institute of Michigan, 1, 161-177. CRAWFORD, G. A. (1995): Environmental improvements by the mining industry in the Sudbury Basin of Canada. J. Geochem. Explor., 52, 267-284. CURRAN, P. J. (1989): Remote sensing of foliar chemistry. Remote Sens. Environ., 30, 271-278. CURRAN, P. J., DUNGAN, J. L., MACLER, B.A. & PLUMMER, S. E. (1991): The effect of a red leaf pigment on the relationship between red edge and chlorphyll concentration. Remote Sens. Environ., 35, 69-76. CURRAN, P. J., DUNGAN, J. L., MACLER, B. A., PLUMMER, S. E. & PETERSON, D. L. (1992): Reflectance spectroscopy of fresh whole leaves for the estimation of chemical concentration. Remote Sens. Environ., 39, 153-166. CURRAN, P. J. & KUPIEC, J. A. (1995): Imaging Spectrometry: A New Tool for Ecology. In: DANSON, F. M. & PLUMMER, S. E. (Eds): Advances in environmental remote sensing. Wiley & Sons, Chichester, 71-88. CURTISS, B. & USTIN, S. L. (1989): Parameters effecting reflectance of coniferous forests in the region of chlorophyll absorption. In: Proc of IGARSS '89, 12th Can Symp on Remote Sensing, Vancouver, BC. DAVIS, A. D. & WEBB, C. J. (1995): Abandoned mines inventory and reclamation in the Black Hills of South Dakota. In: SCHEINER, B. J., CHATWIN, T. D., EL-SHALL, H., KAWATRA, S. K. & TORMA, A. E. (Eds): New remediation technology in the changing environmental arena. Littleton, Colorado. Society for Mining, Metallurgy, and Exploration, Inc., 27-33. DE VOS, K. J., BLOWES, D. W., ROBERTSON, W. D. & GREENHOUSE, J. P. (1995): Delineation and evaluation of a plume of tailings derived water, Copper Cliff, Ontario. Mining and the Environment, Proceedings, Sudbury, 673-682. DIAS, N. W. (2002): Applying Red-Edge & WaterAbsorption Geometry Analyses to Estimated Hardwood Forest Structure and Biomass by Using Hyperspectral Data, EOM, 11, 2.
Environmental Geology, 3 Remote Sensing
155
DOUGLAS, W. J. (1995): Environmental GIS – Applications to industrial facilities. Lewis Publishers, Boca Raton, FL. EHRENBERG, M. (1991): Beprobungslose Altlastenerkundung. wlb Wasser, Luft und Boden, 7-8, 56-59. Ekwan Technology Corporation: www.ekwantech.com ELLIS, J. M., DAVIS, H. H. & ZAMUDIO, J. A. (2001): Exploring for onshore oil seeps with hyperspectral imaging. Oil and Gas Journal, 99, 10, 49-58. ERB, W. & Endres, L. (1989): Leitfaden der Spektroradiometrie. Springer, Berlin. FRICKER, P., SANDAU, R. & WALKER, A. S. (1999): Digital photogrammetric cameras: possibilities and problems. In: FRITSCH, D. & SPILLER, R. (Eds.): Photogrammetric Week ’99, Wichmann Verlag, Heidelberg, 71-82. GAFFEY, S. J., MCFADDEN, L. A., NASH, D. & PIETERS, C. M. (1993): Ultraviolet, visible, and nearinfrared reflectance spectroscopy: Laboratory spectra of geologic materials. In: PIETERS, C. M. & ENGLERT, P. A. J. (Eds.): Remote Chemical Analysis: Elemental and Mineralogical Composition, Cambridge University Press, 43-77. GAMBA, P. and B. HOUSHMAND (2002): Joint analysis of SAR, LIDAR and aerial imagery for simultaneous extraction of land cover, DTM and 3D shape of buildings, International Journal of Remote Sensing, 23, 20, 4439-4450. GATES, D. M., KEEGAN, H. J., SCHLETER, J. D. & WEIDNER, V. R. (1965): Spectral properties of plants. Applied Optics, 4/1, 11-20. GEBHARDT, A. (1981): Thermografie, Anwendungen bei der geophysikalischen Naherkundung. Freiberger Forschungshefte, C367, Dt. Verlag f. Grundstoffindustrie, Leipzig. GIM International (2001): Product Survey on Airborne Laser Scanner - an Overview. GLAESER, C. (1989): Beiträge zur Anwendung der Multispektraltechnik für die Lösung geowissenschaftlicher Aufgaben. Dissertation B, Martin-Luther-Universität, HalleWittenberg, Textband. GOODCHILD, M. F., STEYAERT, L. T., PARKS, B. O., JOHNSTON, C., MAIDMENT, D., CRANE, M. & GLENDINNING, S. (Eds.) (1996): GIS and environmental modeling. Progress and research issues: GIS World Books, Fort Collins, Colorado. GRAHAM, D. F., ST-ARNAND, E. L. & RENCZ, A. N. (1994): Canada Geologic Survey Monitors Mine Tailings, Disposal sites with Landsat. EOS Magazine, 38-41. GUNN, J. M. (ed.) (1995): Restoration and Recovery of an Industrial Region. Springer, New York. GUPTA, R. P. (1991): Remote Sensing Geology. Springer, Berlin. GUPTA, R. P. (2003): Remote Sensing Geology. Springer, Berlin. HAUFF, PH. L. (1993): Spectral Reflectance Properties of Oil and Hydrocarbon-Bearing Rocks and Sediments. Application Note, 1, 1993, Spectral International Inc., Lafayette, CO.
156
3.3 Nonphotographic Imaging
HEIER, H. (1999): Application and markets for digital airborne cameras. In: FRITSCH, D. & SPILLER, R. (Eds.): Photogrammetric Week ’99, Wichmann Verlag, Heidelberg, 43-49. HÖRIG, B., KÜHN, F., OSCHÜTZ, F. & LEHMANN F. (2001): HyMap hyperspectral remote sensing to dedect hydrocarbons. Int. J. Remote Sensing, 22, 8, 1413-1422. HORNSBY, J. K., BRUCE, B. & MACKENZIE-GREIVE, G. (1989): Monitoring Vegetation Regrowth on Placer Mine Tailings, Bonanza Creek, Yukon Territory. In: Proc of Int Symp on Remote Sensing, 2518-2521. HUNT, G. R. & SALISBURY, J. W. (1976): Visible and nearinfrared spectra of minerals and rocks: XII. Sedimentary Rocks. Mod. Geol., 5, 211-217. HUNT, G. R. (1977): Spectral signature of particulate minerals in the visible and near infrared. Geophysics, 42, 501-513. IRVINE, J. M., STAHL, G., ODENWELLER, J., SMYRE, J. L., EVERS, T. K., DALE, H. & KING, A. L. (2000): Thermal Remote Sensing to Detect Buried Waste Material (Oak Ridge, U.S.A.). In: KUEHN, F., KING, T., HOERIG, B. & PETERS, D. (Eds.): Remote Sensing for Site Characterization, Springer Verlag, Berlin, 96-105. JANSEN, W. T. (1994): The Mapping of Mineral Distributions Using Remotely Sensed Hyperspectral Images and Standard Spectral Libraries. Int. Symposium on Remote Sensing and GIS for Site Characterizations – Applications and Standards, ASTM, San Francisco, Jan. 27.-28. 1994. JOHNSON, A. I., PETTERSSON, C. B. & FULTON, J. L. (Eds.) (1992): Geographic information systems (GIS) and mapping – Practices and standards: ASTM, Publication STP, 1126, 346. KERSTEN, T., BALTSAVIAS, E., SCHWARZ, M. & LEISS, I. (2000): IKONOS-2 CARTERRA™ GEO - erste geometrische Genauigkeitsuntersuchungen in der Schweiz mit hochaufgelösten Satellitendaten. VPKOL 8, 2000. KING, D. J. (1993): Digital frame cameras: the next generation of low cost remote sensing sensors. In: Proc of ASPRS Biennial Workshop on Photography and Videography in the Plant Sciences, Logan, Utah. KING, T. V. V. & RIDLEY, W. I. (1987): Relation of the spectroscopic reflectance of olivine to mineral chemistry and some remote sensing implications. J. Geophys. Res., 92, 11457-11469. KING, T. V. V. & CLARK, R. N. (1989): Spectral Characteristics of Chlorites and MgSerpentine Using High-Resolution Reflectance Spectroscopy. J. Geophys. Res., 94, 13997-14008. KING, T. V. V. (Ed.) (1995): Environmental Considerations of Active and Abandoned Mine Lands: Lessons From Summitville, Colorado. USGS Bulletin 2220. 38. KING, T. V. V., CLARK, R. N. & GREGG, A. S. (2000): Applications of Imaging Spectroscopy Data: A Case Study at Summitville, Colorado. In: KUEHN, F., KING, T., HOERIG, B. & PETERS, D. (Eds.): Remote Sensing for Site Characterization, Springer, Berlin, 164-185.
Environmental Geology, 3 Remote Sensing
157
KOKALY, R. F., CLARK, R. N. & LIVO, K. E. (1987): Mapping the biology and mineralogy of Yellowstone National Park using imaging spectroscopy. In: Summaries of 7th Annual JPL Airborne Earth Science Workshop, R.O. Green, ED., JPL 97-21, 1. KOSCHMANN, A. H. (1949): Structural control of the gold deposits of the Cripple Creek district, Teller County, Colorado. USGS Bulletin 955-B, 60. KRONBERG, P. (1985): Fernerkundung der Erde. Enke, Stuttgart. KRUSE, F. A., HAUFF, P. L., DIETZ, J., BROCK, J. C. & HAMPTON, L. R. (1989): Characterization and mapping of mine waste at Leadville, Colorado using imaging spectroscopy. Boulder, Center for the Study of Earth from Space, U of Colorado, EPA Contract No. 68-01-7251, 2. KÜHN, F. & HÖRIG, B. (1995): Environmental Remote Sensing for Military Excercise Places. - Remote Sensing and GIS for Site Characterizations: Applications and Standards, ASTM STP 1279, American Society for Testing and Materials, 5-16. KÜHN, F. & HÖRIG, B. (1995): Handbuch zur Erkundung des Untergrundes von Deponien und Altlasten, 1, Geofernerkundung. Springer, Berlin. KÜHN, F. & OLEIKIEWITZ, P. (1983): Die Nutzung der Multispektraltechnik zur Früherkennung von senkungs- und erdfallgefährdeten Gebieten. Z. angew. Geol. 29, 6, 71-74. KÜHN, F., TREMBICH, G. & HÖRIG, B. (1997): Multisensor Remote Sensing to Evaluate Hazards Caused by Mining. In: Proceedings of the Twelfth International Conference on Applied Geologic Remote Sensing, 17.-19. November, Denver, Colorado, ERIM, I, 425-432. KÜHN, F., TREMBICH, G. & HÖRIG, B. (1999): Satellite and Airborne Remote Sensing to Detect Hazards Caused by Underground Mining. In: Proceedings of the Thirteenth International Conference on Applied Geologic Remote Sensing, 1.-3. March, Vancouver, British Columbia, Canada, ERIM, II, 57-64. KUEHN, F., KING, T., HOERIG, B. & PETERS, D. (2000): Remote Sensing for Site Characterization. Methods in Enviromental Geology. Springer, Berlin. KÜHN, F., MARGANE, A., TATONG, T. & WEVER, T. (2004): InSAR – Based Land Subsidence Map for Bangkok, Thailand. Z. Angew. Geol. 1/2004, 74 -81. LILLESAND, T. M. & KIEFER, R. W. (1994): Remote Sensing and Image Interpretation. 3rd edition, Wiley & Sons, New York. LINTZ, J. & SIMONETT, D. S. (Eds.) (1976): Remote Sensing of Environment. AddisonWesley, London. LIVO, K. E. (1994): Use of remote sensing to characterize hydrothermal alteration of the Cripple Creek area, Colorado. Colorado School of Mines, M. Sc. Thesis T-4613 (unpubl.), Golden. LOWE, D. S. (1969): Optical sensors. In: Principles and applications to earth resources surveys. CNS and Univ. of Michigan, Paris, 73-136.
158
3.3 Nonphotographic Imaging
LYON, J. G. & MCCARTHY, J. (Eds.) (1995): Wetland and environmental applications of GIS. Lewis Publishers, Boca Raton, Florida. LYON, R. J. P. (1994): Weathering and desert varnish in arid terrains, In: Proceedings of the First International Airborne Remote Sensing Conference and Exhibition, Strasbourg, France, Sept. 11.-15., 1994. Ann Arbor, Environmental Research Institute of Michigan, 1, 257-268. MCCARTHY, F., CHENG, P. & TOUTIN, T. (2000): Case Study of Using IKONOS Imagery in Small Municipalities. EOM 11. MCGREGOR, R. G., BLOWES, D. W. & ROBERTSON, W. D. (1995): The Application of chemical Extractions to Sulfide Tailings at the Copper cliff Tailings Area, Sudbury, Ontario. Mining and the Environment, Sudbury 95, Proceedings, Sudbury, 1133-1142. MCCANN, M. E. & KING, A. (1995): Case Study of Advanced Technologies for Hydrological Site Characterization. Mining and the Environment, Sudbury 95 Proceedings, Sudbury, 667-671. MEER, F. D. van der & JONG, S. M. de (Eds.) (2001): Imaging spectrometry: basic principles and prospective applications / Kluwer Academic. MEINEL, G., REDER, J. & NEUBERT, M. (2001): IKONOS- Satellitenbilddaten und ihre Klassifikation – ein erster Erfahrungsbericht. Kartographische Nachrichten, 1, 2000, 40-46. MILLER, V. & MILLER, C. F. (1961): Photogeology. Mc Graw-Hill Book Company Inc. New York, 1961. MILLER, J. R., HARE, E. W., HOLLINGER, A. B. & STURGEON, D. R. (1987): Imaging spectroscopy as a tool for botanical mapping. In: Vane, G. (Ed): Imaging Spectroscopy II, International Society for Optical Engineering, Bellingham, WA, 108-113. MILTON, N. M., AGER, C. M., EISWERTH, B. A. & POWER, M. S. (1989): Arsenic and selenium-induced changes in spectral reflectance and morphology of soybean plants. Remote Sens. Environ., 30, 263-269. MUNTS, S. R., HAUFF, P. L., SEELOS, A. & MCDONALD, B. (1993): Reflectance spectroscopy of selected base-metal bearing tailings with implications for remote sensing. In: Proc of 9th Thematic Conf on Geol Remote Sensing, Pasadena, CA, Feb. 8.–11., 1993: Ann Arbor, Environmental Research Institute of Michigan, 567-578. MUSSOKOWSKI, R. (1983): A Technique for Mapping Environmental Change - using Digital Landsat Data.- COSPAR, Advances in Space Research, 8, 103-107. MURRAY, I. & WILLIAMS, P. C. (1987): Chemical principles of near-infrared. In: WILLIAMS, P. & NORRIES, K. (Eds): Near Infrared Technology in the Agricultural and Food Industries, American Association of Cereal Chemists, St. Paul, MN, 17-37. ONGSOMWANG, S. (1994): Forest Inventory, Remote Sensing and GIS (Geographic Information System) for Forest Management in Thailand. Berliner Geographische Studien, Berlin. PARKINSON, C. L. & MCGUIRE, A. (Eds.) (1997): Using Color-Coded Satellite Images To Examine The Global Enviroment. University Science Books.
Environmental Geology, 3 Remote Sensing
159
PETERS, D. C. & PHOEBE, L. H. (2000): Multispectral Remote Sensing to Characterize Mine Waste (Cripple Creek and Goldfield, U.S.A). In: KUEHN, F., KING, T., HOERIG, B. & PETERS, D. (Eds.): Remote Sensing for Site Characterization, Springer, Berlin, 113-164. PISCHEL, P. & NEUKUM, G. (1999): The High Resolution Stereo Camera HRSC-A – Digital 3-D Image Acquisition; Photogrammetric Processing and Data Evaluation. In: Proceedings, Joint Workshop "Sensors and Mapping from Space 1999", Institut für Photogrammetrie und Ingenieurvermessung, Universität Hannover, 18, 1999. RICHARDS, I. A. (2005): Remote Sensing Digital Image Analysis. 4th ed., Springer, Berlin. ROBERTS, D. A., ADAMS, J. B. & SMITH, M. O. (1993): Discriminating Green Vegetation, Non-Photosynthetic Vegetation and Soils in AVIRIS Data, Remote Sensing of Environment, 44 (2/3), 255-270. ROBERTS, D. A. & HEROLD, M. (2004): Imaging spectrometry of urban materials. In: KING, P., RAMSEY, M.S. and G. SWAYZE, (Eds.): Infrared Spectroscopy in Geochemistry, Exploration and Remote Sensing, Mineral Association of Canada, Short Course Series Volume 33, London, Ontario, 155-181. URL: http://www.ncgia.ucsb.edu/ncrst/research/pavementhealth/urban/imaging_spectr ometry_of_urban_materials.pdf
ROCK, B. A. (1988): Comparison of in-situ and airborne spectral measurements of the blue shift associated with forest decline. Remote Sensing Environ, 24, 109-127. ROWAN, J. S., BARNES, S. J. A., HETHERINGTON, S. L., LAMBERS, B. & PARSONS, F. (1995): Geomorphology and pollution – the environmental impacts of lead mining, Leadhills, Scotland. J. Geochem. Explor., 52, 57-65. SABINS, F. F. (1996): Remote sensing - Principles and interpretation. 3rd. edition, Freeman, San Francisco. SALOMONS, W. (1995): Environmental impact of metals derived from mining activities. Processes, predictions, prevention. J. Geochem. Explor., 52, 5-23. SENGUPTA, M. (1993): Environmental impacts of mining. Lewis, Boca Raton, Florida. SINGHROY, V. H. (1992): Spectral Characterization of the Boreal Forest Species Associated with Geochemical Anomalies. International Geological Congress, Abstracts, Kyoto, Japan, 3, 980. SINGHROY, V. H. (1995): Spectral characterization of vegetation at mine tailings. Mining and the Environment, Sudbury 95 Proceedings, Sudbury, 193-200. SINGHROY, V. H. & KRUSE, F. (1991): Detection of metal stress in the Boreal Forest species using the 670 µm chlorophyll band. ERIM, 8th Thematic Conf on Geologic Remote Sensing, 361-372. SLONECKER, E. T. & WILLIAMS, D. J. (2001): Imaging spectroscopy for detecting fugitive environmental contaminants. Enviromental Protection Agency Report. SMITH, R. B. (2001): Indroduction to Hyperspectral Imaging, MicroImages Inc. STEWART, K. C. & SEVERSON, R. C. (Eds.) (1994): Guidebook on the geology, history, and surface-water contamination and remediation in the area from Denver to Idaho Springs, Colorado. USGS Circular 1097, 55.
160
3.3 Nonphotographic Imaging
STRATHMANN, F.-W. (1993): Taschenbuch zur Fernerkundung. Wichmann, Karlsruhe. STRUHSACKER, D. W. (1995): The importance of waste characterization in effective environmental planning, project design and reclamation. In: SCHEINER, B. J., CHATWIN, T. D., EL-SHALL, H., KAWATRA, S. K. & TORMA, A. E. (Eds): New remediation technology in the changing environmental arena: Littleton, Colorado, Society for Mining, Metallurgy, and Exploration, Inc., 19-25. TANDY, B. C. & AMOS, E. (1985): Airborne thermal infrared linescan in geology. Proc of Intern Symp on Remote Sensing of Environment, 4th Thematic Conf “Remote Sensing for Exploration Geology”, San Francisco, California, 1. - 4. April. TANG, L., DOERSTEL, C., JACOBSON, K., HEIPKE, C. & HINZ, A. (2000): Geometric accuracy potential of the Digital Modular Camera, IAPRS, XXXIII, Amsterdam. THEILEN-WILLIGE, R. (1993): Umweltbeobachtung durch Fernerkundung. Enke, Stuttgart. THOMPSON, T. B., TRIPPEL, A. D. & DWELLEY, P. C. (1985): Mineralized veins and breccias of the Cripple Creek District, Colorado. Econom Geol, 80, 1669-1688. TREVETT, J. W. (1983): Imaging radar for resource surveys. Chapman & Hall, London TOUTIN, T. & CHENG, P. (2000): Demystification of IKONOS, EOM 9, 7. U.S. Forest Service (1993): Acid drainage from mines on the national forests: a management challenge. Program Aid 1505, 12. USTIN, S. L., MARTENS, S. N., CURTISS, B. & VANDERBILT, V. C. (1994): Use of high spectral resolution sensors to detect air pollution injury in conifer forests. In: FENSTERNMAKER, L. A. (Ed.): Remote sensing applications for acid deposition. EPA publication CR81400201, 72-85. VANE, G., GREEN, R., CHRIEN, T., UNMARK, H., HANSEN, E. & PORTER, W. (1993): The airborne visible/infrared imaging spectrometer (AVIRIS). Remote Sensing Environ., 44, 127-143. VANGRONSVELD, J., STERCKX, J., VAN ASSCHE, F. & CLIJSTERS, H. (1995): Rehabilitation studies on an old non-ferrous waste dumping ground. Effects of revegetation and metal immobilization by beringite. J. Geochem. Explor., 52, 221-229. VARSHNEY, P. & ARORA, M. (2004): Advanced Image Processing Techniques for Remotely Sensed Hyperspectral Data. Springer, Berlin. VEGT, J. W. & HOFFMANN, A. (2001): Airborne Laser Scanning Reaches Maturity. Geoinformatics, Sept. 2001, Emmelord, 32-39. WOODING, M. G. (1988): Imaging radar applications in Europe, illustrated experimental results (1978-1987). ESA TM-01, Noordwijk. ZILIOLI, E., GOMARASCA, M. A., & TOMASONI, R. (1992): Application of terrestrial thermography to the detection of waste-disposal sites. Remote Sensing Environ, 40, 2, 153-160.
Environmental Geology, 4 Geophysics
161
4 Geophysics NORBERT BLINDOW, KLAUS KNÖDEL, FRANZ KÖNIG, GERHARD LANGE, HARALD LINDNER, REINIE MEYER, KLAUS-HENRIK MITTENZWEY, ANDREAS SCHUCK, KNUT SEIDEL, PETER WEIDELT, THOMAS WONIK, UGUR YARAMANCI, with contributions by DIETER EISENBURGER, BERNHARD ILLICH, RICARDO A. OLEA, HELLFRIED PETZOLD & THOMAS RICHTER
4.1 Magnetic Methods KLAUS KNÖDEL
4.1.1 Principle of the Methods Everywhere on the Earth there is a natural magnetic field which moves a horizontally free-moving magnetic needle (magnetic compass) to magnetic north. The magnetic field is a vector field, i.e., it is described by its magnitude and direction. The magnetic field consists of three parts: the main field, a fluctuating field, and a local anomaly field.
Fig. 4.1-1: Typical applications of the magnetic method, left: search for concealed waste disposal sites containing magnetic materials, right: investigation of the geological structures
162
4.1 Magnetic Methods
The main field, whose origin is within the Earth, varies very slowly with time (years to decades) and is superimposed by a rapidly varying (in fractions of seconds to days) field component, whose origin is outside the Earth (external field, time-varying field). In addition to these geomagnetic field components, there is an almost constant local anomaly field 'F, which results from the magnetization of material in the upper crust. Not only do geological bodies, consisting of, for example, basalt, metamorphic rocks, and some ore deposits, cause local magnetic anomalies, but also metal objects at legal and illegal waste disposal sites (Fig. 4.1-1). The values of the magnetic flux density are 25 000 to 60 000 nT for the main field, 0.1 to several 100 nT for the fluctuating field, and up to several 1000 nT for the local anomaly field. The local anomaly field is estimated from magnetic data and conclusions are made about the sources. To obtain the local anomaly field, the main field and the time-varying field must be eliminated from the measured total field. The local anomaly field is represented on an isoline map, (pseudo 3-D plot) or along profiles. The sources of the magnetic anomalies are interpreted from these maps and, in some cases, from model calculations for two and/or threedimensional models.
4.1.2 Applications x
Search for concealed waste disposal sites containing magnetic materials,
x
localization of concentrations of barrels hazardous waste in a landfill,
x
determination of the parts of a landfill with a high proportion of building rubble,
x
search for unexploded ordnance (UXO),
x
investigation of the geology below planned landfills and industrial or mining sites in areas of igneous and metamorphic rocks, and
x
delineation of faults, particularly in areas of igneous and metamorphic rocks.
suspected of containing
4.1.3 Fundamentals The International Association for Geomagnetism and Aeronomy in 1973 resolved that geomagnetic field measurements shall be reported in tesla, i.e., the SI units (International System of units) of magnetic flux density B. One tesla (T) is equal to 109 nanotesla (nT). Although all magnetometers are now
Environmental Geology, 4 Geophysics
163
calibrated in units of magnetic flux density, the term “magnetic field” has been retained for the measured quantity. For this reason, the strength of the total magnetic field, the time-varying field, or the local anomaly field is always understood as magnetic flux density. In air, the relationship between magnetic flux density B and magnetic field strength H is approximated as follows: B | µ 0H. The most widely used magnetometers in applied geomagnetics measure the intensity F of the total field F. 'F is the intensity of the local anomaly field, where 'F = F - F0. 'F is a good approximation of the projection of 'F on the main field F0. The local magnetic anomaly field is due to local and regional differences in the magnetization of rocks (Table 4.1-1) and other magnetic materials. A distinction is made between induced (i) and remanent (r) magnetization. These add vectorially: M = Mi + Mr. The induced magnetization Mi is proportional to the magnetic field strength H:
Mi = N H = N B / µ 0
(4.1.1)
B = µ 0 µ H = µ 0 (1 + N) H = µ 0 (H + Mi)
(4.1.2)
In magnetically anisotropic material µ and N are tensors. The magnetic susceptibility N and the relative magnetic permeability µ are properties of magnetic materials (Fig. 4.1-2). The susceptibility is sometimes required in cgs units for model calculation programs: Ncgs { (1/4S)NSI.
Fig. 4.1-2: Ranges of values for the magnetic susceptibility N of different rocks, compiled by SCHÖN (1983)
164
4.1 Magnetic Methods
According to BREINER (1973), the magnetic susceptibility of most iron and steel objects is between 10 and 130 (SI). Stainless steel is practically nonmagnetic. The magnetic susceptibility of corroded scrap metal is 40 000 - 750 000 10-6 (SI). Table 4.1-1: Magnetization (in A m-1) of the most important rock types, Bosum (1981)
Mi
mean Mi
Mr
mean Mr
Ultrabasic igneous rocks pyroxenite serpentinite
0.01 - 0.45 0.5 - 10
0.1 - 0.2 0.75
0.2 - 0.8
0.6
Basic igneous rocks basalt diabase gabbro
0.01 - 100 0 - 3.5 0.05 - 3
0.02 - 1.5 0.04 - 0.9
0.01 - 200 0-4 0.2 - 30
0.05 - 3 0.05 - 0.9
Intermediate and acidic igneous rocks diorite porphyry granite
0 - 1.5 0 - 0.3 0 - 2.2
0.03 - 0.5
0-9
Metamorphic rocks amphibolite gneiss
0 - 1.3 0 - 1.8
0.01 - 0.75
0 - 1.6 0-2
Sedimentary rocks
0 - 0.4
Magnetic ores magnetite pyrrhotite hematite chromite
20 - 1000 0.3 - 50 0 - 0.9 0.01 - 2
0 - 0.003
0.01 - 1.4
0 - 0.4
20
30 - 1000 2 0.1 - 0.25 0.02 - 30
Conventional methods yield an apparent susceptibility N of the source body as a whole, which is related to the susceptibility NM of the magnetic substances in that body as follows: N = (p NM) / (1 + N NM), where p is the proportion of magnetized grains in the rock and N is a demagnetization factor. N is dependent on the shape of the grains: It equals 1/3 for homogeneous, magnetically isotropic spheres. The magnetization of a rock or material that remains in the absence of an external field is called remanent magnetization, Mr, which depends on the content of ferrimagnetic matter and on the formation conditions.
Environmental Geology, 4 Geophysics
165
A sphere with homogeneous magnetization is the simplest model of a magnetic geological or artificial body. The magnetic field (i.e., magnetic flux density) of a homogeneously magnetized sphere is equal to the field of a dipole at the center of the sphere when the distance r between source and point of measurement is larger than the radius R and the magnetic moment m = (4S / 3) R³ M = V M, where V is the volume of the sphere. From this, it follows that for all spheres for which R < r and which have the same R³ M value and the same center, the same magnetic anomaly is observed (equivalence principle of potential theory). If (a) the center of a sphere with a radius R lies at a depth d (d > R), (b) I is 1 the inclination of the magnetic field at this site, (c) D is the angle between magnetic north and the x-axis, and (c) the origin (x = y = z = 0) of the geophysical coordinate system lies directly above the center of the sphere, then the anomaly of the total field GF can be estimated as follows (LINDNER & SCHEIBE, 1978): GF ( x)
4 SR 3țF ½ 2 2 2 2 2 2 2 2 °3 0 ° 3d sin I 3 x cos I cos Į 3 xd sin I cos Į x d (4.1.3) ® ¾ (x 2 d 2)5 / 2 °¯ 4S °¿ factor
Taking the magnetic anomaly at the Earth's surface above the center of the sphere at x = 0 into consideration, it can be seen that the magnetic field directly above the source body decreases with the third power of the depth d. The local anomaly field depends not only on the size and magnetization of the source body but also to a large degree on its depth. Small near-surface source bodies in landfills cause a strongly varying local anomaly field. These local variations smooth rapidly with increasing height h above the surface (1 to 4 m). For further formulas to calculating GF anomalies of model bodies with a simple geometry, see TELFORD, GELDART & SHERIFF (1990). Figure 4.1-3 shows the typical appearance of the anomalies of the total field intensity for inclinations I = 15o and 30° (e.g., South East Asia) and I = 67.5o (e.g., Central Europe). These anomalies, which are due to induced magnetization, are superimposed by the anomaly field due to the remanent magnetization. The measured value 'F is the sum of these two local anomalies. The anomalies of the total magnetic field can be calculated using Equation (4.1.3.) To estimate the maximum value expected for the anomaly, set x = 0 and I = 90° in Equation (4.1.3.) This yields
1
The inclination is the angle between the vector of the main magnetic field and the horizontal.
166
GFmax
4.1 Magnetic Methods
3 S R 3 2N F0 4 . 4S d 3
(4.1.4)
Introducing the parameter “effective susceptibility” Neff, we can define the magnetic moment mc as follows: 3 S N eff R 3 F0 . (4.1.5) 4 For a spherical body at a depth d, the maximum value of the anomaly of the total field intensity is mc
GFmax
mc . 4S d 3
(4.1.6)
Mean values for the normal magnetic field in the Berlin area are F0 = 49 000 nT and I = 67.5°. Values of N = 100 (SI) can be assumed for the magnetic susceptibility of iron and steel and 6600 - 8100 kg m-3 (mean 7800 kg m3) for their density. Figure 4.1-4 shows a nomogram for Equation (4.1.6). This nomogram shows that at the Earth’s surface the local anomaly field of a model body containing 100 kg iron at a depth of 5 m has an maximum value of 50 nT.
Fig. 4.1-3: Anomaly (induced part) of the total field intensity for a sphere with a radius R and a homogeneous magnetization, at x = 0, at a depth d = 10 m, for inclination I = 15o, 30o, 67.5o and D = 0
167
Environmental Geology, 4 Geophysics
Fig. 4.1-4: Nomogram for estimating the maximum anomalies GFmax from iron objects with mc = 5·4S·10-8 Wb m per kilogram after BREINER (1973); taken from MILITZER et al. (1986). Note that GFmax can change by up to a factor 5, depending on the hardness of the iron, the inclination, and the direction of the remanent magnetization vector
Table 4.1-2 can be used to estimate the magnetic anomaly of barrels before the measurements are carried out. Table 4.1-2: Size and mass of steel containers for waste disposal and the maximum magnetic fields GFmax according to Equation (4.1.6) at a depth d. The magnetic moment mc = 5·4S·10-8 Wb m per kilogram Local anomaly field GFmax in nT d=1m d=3m d=5m capacity outside diameter length sheet metal thickness mass (empty)
30 - 220 L 290 - 610 mm 440 - 950 mm 0.5 - 1.2 mm 2.6 - 22.6 kg
130 - 1130
5 - 42
1-9
168
4.1 Magnetic Methods
4.1.4 Instruments Geomagnetic fields are measured with fluxgate magnetometers, protonprecession magnetometers, and optically pumped magnetometers. Fluxgate magnetometers measure the components of the magnetic field (e.g., vertical intensity) and its gradients. Proton-precession magnetometers and optically pumped magnetometers measure total field intensity. Fluxgate magnetometers utilize the nonlinearity of the magnetization curve B as a function of H of the high permeability magnetic material of the coil core. The core is magnetized to saturation by an ac field in order to measure the component of the magnetic field that is parallel to the core axis. Continuous measurement is possible. At a walking pace, 1 - 8 readings can be taken per profile meter. Thus, the high density of measurement points necessary for environmental studies can be obtained within an acceptable time. Fluxgate magnetometers measure 1 or 3 components of the geomagnetic field or they measure the vertical gradient of the vertical component. State-ofthe-art fluxgate magnetometers have the following specifications: sensitivity
0.1 nT
precision
±1 nT
dynamic range
±2000 or 20 000 nT
response time
20 ms
temperature range
-10 to +40 °C
temperature drift
0.1 nT/°C.
The most widely used magnetometers in applied magnetics are protonprecession magnetometers. They measure the precession frequency of protons in the geomagnetic field after a polarization field is switched off. The precession frequency is proportional to the field strength of the total magnetic field. The Overhauser effect (transfer of the spin moment of electrons to protons by electromagnetic waves with a frequency selected for a specific chemical compound) is used to overcome the disadvantage of discontinuous readings of the proton-precession magnetometers. State-of-the-art proton-precession magnetometers have the following specifications: sensitivity precision dynamic range gradient tolerance response time temperature range
0.1 nT ±1 nT 18 000 to 110 000 nT 6000 nT/m 0.5 to 3 s -40 to +55 °C.
Environmental Geology, 4 Geophysics
169
Almost all proton-precession magnetometers have gradient probes for measuring the vertical gradient of the total field. The distance 'r between the sensors for the gradient measurement is 0.5 - 2 m. 'r must be less than 1/5 - 1/10 the distance r of the gradient probes from the source body (BREINER, 1973). This condition is certainly fulfilled when geological structures are being explored, but not always for investigations of landfills. Not all gradient probes measure the gradients simultaneously by both sensors. The measurement of the total fields by the two sensors 1 - 2 s apart can lead to incorrect gradient measurements in areas where the magnetic field changes rapidly or if there is movement of the probes, for instance, by the wind. Gradiometers have the following advantages: higher sensitivity with respect to near-surface targets and elimination of the fluctuating part of the geomagnetic field and regional trends. Proton-precession magnetometers with a response time of 0.5 s make essentially continuous measurements possible if the magnetometer is carried at walking speed along a profile. Proton-precession magnetometers are also available as base stations for recording variation of the total field. The time between two readings can be chosen between 5 s and 60 min. Optically pumped magnetometers (also called alkali vapor magnetometers or optical absorption magnetometers) are based on the Zeeman effect, the splitting of the spectral lines in the Earth’s magnetic field. The distance between the split spectral lines is proportional to the magnetic field strength. The optical pumping technique is used instead of spectrometric methods. Rubidium, caesium or potassium vapor is used in optical absorption cells. Therefore, such magnetometers are referred to as rubidium or caesium, or potassium vapor magnetometers. The sensitivity of this type of magnetometer is about two orders of magnitude higher than that of proton-precession magnetometers. This sensitivity is needed only for special investigations. The more sensitive optically pumped magnetometers, with which measurements can be made more rapidly, are more expensive than proton-precession magnetometers. All modern magnetometers give direct readings of the magnetic field values in nanoteslas. A total of 10 000 - 100 000 readings can be stored in a datalogger for subsequent downloading to a computer. Nowadays magnetometers integrated with differential GPS (Global Positioning Systems) are available, which makes it unnecessary to use a regular survey grid (TAMIR L KLAFF, 1999). Particularly for unexploded ordnance (UXO) investigations, the combined use of magnetic and electromagnetic sensors integrated with differential GPS can be useful (BARROW et al., 1996).
170
4.1 Magnetic Methods
4.1.5 Survey Practice The objectives must be known in detail for optimum planning and execution of a magnetic survey. Also of importance is knowledge of the local geology and the situation in the field (e.g., accessibility, benchmarks, sources of noise). The parameters and equipment for the survey are chosen on the basis of this information: 1. Planning of the survey grid: a) location and size of the study area (possibly on the basis of test measurements), b) line spacing, c) spacing of measurement points, d) orientation of the profiles, e) consideration of utility lines (water, gas, electricity, telephone), f) choice of an undisturbed base point and if necessary further reference and control points, g) minimum number of points for replicate measurements, h) sequence of measurements. 2. measured parameter (e.g., total field intensity and/or vertical gradient of the total field intensity, vertical gradient of the vertical component), 3. precision, 4. distance of the sensors from the Earth's surface, 5. use of a base station, 6. type of magnetometer. The survey area should be large enough that the non-anomalous area surrounding the target area is also included. For landfills, the area surveyed must be larger than the planned landfill. In the normal case, magnetic measurements are carried out along profiles. The profiles should be laid out perpendicular to the strike of the geological structures. If the strike is known, a line spacing of 2 to 5 times the distance between measurement points is often used. Grids with same spacing for profiles and measurement points lead to the most reliable results. If the magnetic field fluctuates irregularly and the line spacing is too large, the anomalies will indicate an erroneous strike direction. An example of this is given by HAHN et al. (1985) (pg. 126), who recommends that the spacing of profiles and measurement points be about one-third of the expected depth of the target(s). For investigations of landfills, the grid spacing should be 5 - 10 m. The spacing can be reduced for more detailed measurement of
Environmental Geology, 4 Geophysics
171
source bodies of interest and for exact determination of the boundaries of concealed landfills. The search for concealed barrels normally requires a grid spacing of less than 1 m. In such cases, quasi-continuous measurements with two readings per second or per meter are useful. Measurement of the total field anomaly 'F is often impossible near electric railways. However, simultaneous sensor readings for gradient measurements still produce useful results in this case. Iron water pipes disturb gradient measurements more strongly than 'F measurements (LINDNER et al., 1984). Therefore, particularly in areas in which there are magnetic disturbances, it is recommended that the total field and gradient be measured simultaneously. Since the local magnetic anomaly field decreases approximately with the third power of the distance of the probe from the target, the influence of small near-surface bodies decreases quickly with increasing height of the sensor above the Earth’s surface. This effect is used to eliminate the anomalies resulting from these near-surface bodies and to smooth the measured local anomaly fields. Depending on the objective of the survey, the sensors are placed 0.3 - 4 m above the Earth’s surface. To avoid interference fields t 1 nT, a distance of at least 30 m should be kept from motor vehicles and 150 m from the steel towers of power lines (HAHN et al., 1985). See Section 4.1.9 for examples of parameter values for various survey objectives.
4.1.6 Processing and Interpretation of the Measured Data The gradient values do not require any further processing after measurement (i.e., main field and variations are already eliminated). But the following corrections of the total field must be carried out to determine the local anomaly field 'F: Corrections for diurnal variations: Variations of the geomagnetic field are best measured at a base station within the survey area which automatically records the total field, for example, every 10 s. These values can be used to eliminate the time-varying field from all measurements. The time-varying field can also be largely eliminated by frequent control measurements (tie-line method). Normal field and elevation corrections: The normal field can be described as a function of geographical longitude and latitude, altitude above sea level and time (e.g., January 1994 = Epoch 1994.1). Tables of the normal field or computer programs using the International Geomagnetic Reference Field (IGRF) are employed to calculate the normal field. Owing to irregular secular variations of the main field, the IGRF is updated every five years. Because the survey area for a site investigation is normally small, the differences in measurement elevation are small and the measurements are
172
4.1 Magnetic Methods
made over a short period of time, one normal-field value F0 can be used for the entire survey area. A single base station in a neighboring undisturbed area is also sufficient. Terrain correction is not necessary for the objectives discussed here. In special cases, a drift correction is necessary for fluxgate magnetometer measurements. For qualitative investigations, such as the search for and delineation of abandoned industrial and landfill sites, a correction for diurnal variation or for normal field and elevation is not often done, because anomalies caused by abandoned landfills also are clearly visible in the total field. The anomalies of the total field and/or of the vertical gradients are represented on a contour map and/or a set of profiles. Color or grey shading increase the readability of contour maps. Three-dimensional plots are normally suitable only for a general impression of the results. The delineation of underground structures is too inaccurate with pseudo 3-D maps. Trend removal, filtering, field continuation, second-derivative analysis, Fourier analysis, pole reduction, and reduction to the magnetic equator can be helpful for the interpretation of the data, making the underground structures more clearly visible on the anomaly maps. For details, see TELFORD et al. (1990). Information about the subsurface can often be derived directly from isoline maps and profiles by qualitative interpretation by a geophysicist. Some guidelines to qualitative interpretation of magnetic profiles and maps are listed in Table 4.1-3. The geometry, depth, magnetic susceptibility (and remanent magnetization) of the magnetic material (e.g., geological structures, wrecked cars or barrels, UXOs) can be estimated by quantitative interpretation. Magnetic effects of simple bodies are depicted, for example by TELFORD et al. (1990) and REYNOLDS (1997). These depictions are useful in order to get an understanding of possible causes of magnetic anomalies. As rule of thumb, the half-width of the anomaly corresponds approximately to 1 to 3 times the depth of the center of the model body. The depth d of a model body can also be estimated from measurements of the total field anomaly 'F and the vertical gradient dF/dz (BREINER, 1973). The following relation is valid at the maximum of the anomaly: d = (-n'F) / (dF/dz), where n = 3 for dipole sources, n = 2 for a horizontal cylinder, and n = 1 for a narrow vertical dike. For this equation, only an assumption about the type of model body is necessary and not about the magnetic susceptibility. Other simple methods for depth estimation of magnetically effective structures in the subsurface as Peters‘ Half-Slope method (PETERS, 1949) and Parasnis‘ method (PARASNIS, 1986) are described in REYNOLDS (1997). Forward modeling and inversion programs for 2, 2.5, and 2.75-D magnetic and/or gravity models are current practice. Three-dimensional structures can be modeled in special cases. Model calculations can also take into account the
173
Environmental Geology, 4 Geophysics
remanent magnetization. Inversion of measured data cannot unambiguously yield the correct model, owing to the principle of equivalence. The number of possible models must be reduced on the basis of geological knowledge and/or by comparison with the results of other methods. In any case, it is essential that the field data used for modeling is adequate with respect to data density and quality. Table 4.1-3: Guidelines to qualitative interpretation of magnetic profiles and maps, modified after REYNOLDS (1997). © John Wiley and Sons Ltd. Reproduced with permission. Taken into consideration segments of a profile and areas of maps
anomaly
Observation
Indication for
magnetically quiet field
near surface rocks with a low magnetic susceptibility N
magnetically noisy field
near surface rocks with a moderate to high magnetic susceptibility N
Wavelength
short => near-surface feature long => deep-seated feature
profiles and maps
positive or negative amplitude
intensity of magnetization
anomaly structure* and shape
dip and dip direction induced magnetization if the anomaly shows a minimum to the north and a maximum to the south in the northern hemisphere and vice versa in the southern hemisphere; if this is not the case, it implies significant remanent magnetization present
profiles and maps
magnetic gradient
possible contrast in N and/or magnetization direction
maps
linearity in anomaly
possible strike of magnetic feature
dislocation of contours
lateral offset by fault
broadening of contour interval
downthrow of magnetic rocks
*i.e., positive peak only, negative peak only or doublet positive and negative peaks Recent developments in interpretation of magnetic data are target characterization algorithms designed for determination of an optimum source location of geologic structures as well as location and apparent size of buried
174
4.1 Magnetic Methods
ferrometallic objects, such as drums, pipes and UXOs (REID et al., 1990; YAGHOOBIAN et al., 1993; REYNOLDS, 1997). The most common technique is the Euler deconvolution. This method is both a geometry finder and a depth estimator. Euler’s homogeneity equation relates the magnetic field and its gradient components to the location of magnetic sources with the degree of homogeneity expressed as a structural index. The structural index is a measure of the rate of change with the distance of the field from the source and is directly related to the source dimension. The advantages of the Euler deconvolution are that it can be applied to large gridded data sets, no particular geological model is necessary, and Euler’s equation is insensitive to magnetic inclination, declination and remanence. Geologic constraints are imposed by choice of the structural index. In addition to the Euler deconvolution, a 3-D analytic signal is used for magnetic interpretation (ROEST et al., 1992; REYNOLDS, 1997). The analytic signal can be used to estimate magnetic contrast and approximate depth, while the Euler deconvolution can give a more detailed structural interpretation at depth. The absolute value of the analytic signal is defined as the square root of the squared sum of the two horizontal and the vertical derivatives of the total magnetic field. Maxima of the analytic signal indicate locations of magnetic contrasts. The depth of a magnetic source can be estimated from the shape of the analytic signal. The objective of a magnetic survey for investigating a site must be a map which summarizes the results for geophysicists as well as for nonspecialists in a clear and understandable form.
4.1.7 Quality Assurance The general aspects of quality assurance are given under this term in the glossary. The following measures are important for quality assurance in a magnetic survey: x Check that the survey personnel are not carrying any magnetic material and that the sensor is not dirty. x Check direction dependence of the readings (“heading effects”), i.e., with the sensor always held by the technician on the same side of his/her body, making readings in all four cardinal directions. x Check for measurement reproducibility and noise: Equipment check, e.g., for low battery voltage, loose sensor cables, low sensor fluid level, exceedance of gradient tolerance, and sources of noise (from electric railways, power lines, etc.). x Record observations in the area, e. g., sources of disturbance (steel towers for power lines and iron fences).
175
Environmental Geology, 4 Geophysics
x Plot the readings at the end of a field day so that errors and disturbances can be recognized and, if necessary, measurements can be repeated. x Take replicate measurements: should be repeated.
Approximately 2 - 5 % of the readings
x Recognized errors must be removed by repeating the measurements. Obscuring errors by “data processing” is inadmissible.
4.1.8 Personnel, Equipment, Time Needed Personnel mobilisation and demobilisation
Equipment
depends on the distance to the survey area
topographic survey measurements
Time needed
s. Chapter 2.5 1 geophysical technician (+1 assistant)
1 4WD vehicle, 1 magnetometer or gradiometer, (1 base station), 1 PC
a=1-5m L = 800 - 2500 N/d a = 10 m L = 600 - 900 N/d quasi continuous measurement of the gradient with 1 - 8 readings/s L = up to 15 000 N/d
data processing, interpretation, reporting
1 geophysicist (+ 1 assistant)
1 PC with plotter, printer, and software
1 - 2 days are necessary for each day in the field
a = grid spacing, N = number of readings, d = 10-hour work day, L = N/d
176
4.1 Magnetic Methods
4.1.9 Examples Search for and delineation of an concealed landfill A pit about 8 m deep, dug to obtain material for a freeway, was filled with waste until 1979. The landfill was then covered with a layer of soil 1 m thick. Today, the landfill is not visible in the field. The abandoned landfill is in a ca. 50 m thick layer of Triassic sandstone. The magnetic method was used to determine the exact location of the landfill. Survey parameters: size of the study area
400 u 400 m
line spacing
5m
spacing of the measurement points
5m
number of measurement points
6581
duration of the survey
9 days
personnel
1 geophysical technician
physical parameters
total field intensity and vertical gradient
distance of the sensors from the ground surface
2.0 and 1.3 m
equipment
G 856-AX
base station magnetometer
G 856 (sampling rate 10 s).
The concealed landfill stands out clearly in the anomalies of both total field intensity 'F (Fig. 4.1-5) and vertical gradient (Fig. 4.1-6). The delineation of the landfill area is clearest in the isoanomaly plot of the vertical gradient (Fig 4.1-6). Calculations based on the 'F values yielded depths of less than 5 m for the magnetic materials and susceptibility values indicative of iron. There is a high-voltage power line along the 400E grid line with a steel tower at 400E, 520N. The vertical gradients are more strongly disturbed by the power line than the 'F isoanomalies. The anomalies in the eastern part of the survey area are caused by building rubble containing iron. Figure 5.4-25 shows the modeling of the magnetic anomaly related to the Nong Harn landfill near Chiang Mai, Thailand.
Environmental Geology, 4 Geophysics
177
178
4.1 Magnetic Methods
Lithology and tectonic structure below a landfill in an area of igneous and metamorphic rocks The site is in the “outer schist belt” of the Granulite Mountains in Saxony, Germany. The phyllites, hornblende phyllites to hornblende schists strike approximately N - S dipping 20 - 50° to the east. There are local bodies of carbonate rocks. Loess loam of variable thickness locally overlies these rocks. Survey parameters: size of the study area
1400 m u 2000 m
line spacing
10 m
spacing of measurement points
10 m
number of measurement points
26 854
duration of the survey
33 days
personnel
2 geophysical technicians, 1 geophysicist
number of measurements at repetition points and check points
447
physical parameters
total field intensity and vertical gradient
standard error in the field 'F at 2.05 m above ground level
2.3 nT
standard error for the vertical gradient
2.2 nT/m
distance of the sensors from the ground surface
0.65 and 2.05 m
equipment
G 856-AX
base station magnetometer
G 856 (sampling rate 10 s)
The map of the anomalies of the total field (Fig. 4.1-7) shows linear and concentric structures, most of which can be correlated with geological sources. These sources can be subdivided into (i) concealed geological bodies (probably consisting of gneissic mica schists, serpentinites, amphibolites, and phyllites) that have no clear relationship to near-surface structures, (ii) nearsurface rocks and magnetic intrusive rocks, and (iii) outcrops of magnetic material. The parameter values calculated for the model bodies are given in Fig. 4.1-8. Furthermore, expensive studies of the geological barrier can be targeted more precisely, reducing the cost of such investigations.
Environmental Geology, 4 Geophysics
179
180
4.1 Magnetic Methods
Search for unexploded ordnance (UXO) Magnetic and electromagnetic techniques have been used for many years to detect unexploded ordnance concealed in the subsurface. The use of both methods for the same investigation site leads to more reliable and less ambiguous results. In the past “mag and flag” methods, i.e., magnetic measurements without data collection and indications flagged for excavation, were commonly used. Nowadays, surveys with data collection, sophisticated processing and interpretation are state-of-the-art. On the UXO test field of the Berlin police department, different types of UXO and other targets, such as steel girders, a steel ball, and metal containers were buried. Measurements with a magnetic gradiometer (Fig. 4.1-9) and with a transient electromagnetic probe (Fig. 4.1-10) reveal the concealed targets. In addition to the location and depth, the weight of the targets was estimated. In this example, depth estimations on the basis of magnetic anomalies are more reliable than those from EM 61 data. Due to remanent magnetization, the estimated mass is often too small. Survey parameters: size of the study area
25 m u 50 m
25 m u 50 m
line spacing
0.50 m
0.50 m
spacing of measurement points
0.25 m
0.195 m
number of measurement points
10 250
13 100
duration of the survey
1 day
1 day
personnel
1 technician, 1 geophysicist
1 technician, 1 geophysicist
parameters
vertical gradient of vertical component of the magnetic field
electromagnetic response
equipment
FM 36
EM 61
181
Environmental Geology, 4 Geophysics 50
N 10
45 9 40 8 35 7
30 6
25 5 20 4
15 3
10 2
5 1 0
0
5
10
15
20
25 Objects: ID # estimated depth (m) estimated weight (kg)
Fig. 4.1-9: UXO test field of the Berlin police department, vertical gradient of the vertical component of the magnetic field [in nT/m], including interpretation, courtesy of Büro für Geophysik Lorenz, Berlin by permission
182
4.1 Magnetic Methods
50
N 10 45
9 40
8 35
7 30
6 25
5 20
4 15
3 10
2 5
1 0
0
5
10
15
20
25 Objects: ID # estimated depth (m) estimated weight (kg)
Fig. 4.1-10: UXO test field of the Berlin police department, electromagnetic response of EM 61 [in mV], including interpretation, courtesy of Büro für Geophysik Lorenz, Berlin by permission
183
Environmental Geology, 4 Geophysics
Parameters and units Symbol
Units
magnetic field strength
H
A m-1
magnetic flux density or magnetic induction
B
tesla (T) = V s m-2 1 nT = 10-9 T 1 T = 1 Wb m-2
magnetic permeability in a vacuum
µ0
µ 0 = 4S 10-7 V s A-1m-1
relative magnetic permeability
µ
dimensionless
magnetic susceptibility
N
dimensionless
total magnetic field
F
nT
main field (also called normal field, reference field)
F0
nT
local anomaly field
'F
nT
inclination
I
degrees
magnetization
M
A m-1
magnetic dipole moment
m
A m²
magnetic moment after Coulomb
mc
Wb m
References and further reading BARROW, B., KHADR, N., DIMARCO, R. & NELSON, H. H (1996): The Combined Use of Magnetic and Eletromagnetic Sensors for Detection and Characterization of UXO. SAGEEP 1996, 469-477. BOSUM, W. (1981): Anlage und Interpretation aeromagnetischer Vermessungen im Rahmen der Erzprospektion. Geol. Jb., E 20, 3-63. BREINER, S. (1973): Applications manual for portable magnetometers. GeoMetrics, Sunnyvale, CA. HAHN, A., PETERSEN, N. & SOFFEL, H. (1985): Geomagnetik, In: BENDER, F. (Ed.): Angewandte Geowissenschaften, II: Methoden der Angewandten Geophysik und mathematische Verfahren in den Geowissenschaften, Enke, Stuttgart, 57-155. LINDNER, H. & SCHEIBE, R. (1978): Die Berechnung von Gg - und GT -Anomalien für regelmäßige homogene Störkörper. Gerlands Beitr. Geophys., 87, 29-45. LINDNER, H., MAURITSCH, H., MILITZER, H., RÖSLER, R., SCHEIBE, R., SEIBERL, W., WALACH, G. & WEBER, F. (1984), In: MILITZER, H. & WEBER, F. (Eds.): Angewandte Geophysik, 1: Gravimetrie und Magnetik, Springer, Vienna, and Akademie-Verlag Berlin. MILITZER, H. & SCHEIBE, R. (1981): Grundlagen der angewandten Geomagnetik. Freiberger Forschungsheft C 352, VEB Dt. Verlag für Grundstoffindustrie, Leipzig.
184
4.1 Magnetic Methods
MILITZER, H., SCHÖN, J. & STÖTZNER, U. (1986): Angewandte Geophysik im Ingenieur- und Bergbau, 2. Aufl., Enke, Stuttgart. PARASNIS, D. S. (1986): Principles of Applied Geophysics, 4th ed., Chapman & Hall, London. PETERS, L. J. (1949): The direct approach to magnetic interpretation and its practical application. Geophysics, 14, 290-320. REID, A. B., ALLSOP, J. M., GRANSER, H., MILLETT, A. J. & SOMERTON, I. W. (1990): Magnetic interpretation in three dimensions using Euler deconvolution. Geophysics, 55, 80-91. REYNOLDS, J. M. (1997): An Introduction to Applied and Environmental Geophysics. John Wiley & Sons, Ltd., Chichester. ROEST, W. R., VERHOEF, J. & PILKINGTON, M. (1992): Magnetic interpretation using the 3-D analytic signal. Geophysics, 57, 116-125. SCHÖN, J. (1983): Petrophysik, Enke, Stuttgart. TAMIR L KLAFF, P. G. (1999): The Application of an Integrated Magnetometer/Global Positioning System for an Unexploded Ordnance Investigation. SAGEEP, 793-801. TELFORD, W. M., GELDART, L. P. & SHERIFF, R. E. (1990): Applied Geophysics, 2nd ed., Cambridge University Press, Cambridge, 62-135. YAGHOOBIAN, A., BOUSTEAD, G. A. & DOBUSH, T. M. (1993): Object Delineation Using Euler’s Homogeneity Equation, Location and Depth Determination of Buried FerroMetallic Bodies. SAGEEP, 613-632.
Environmental Geology, 4 Geophysics
185
4.2 Gravity Methods KNUT SEIDEL & HARALD LINDNER
4.2.1 Principle of the Methods Gravity is defined as the force of mutual attraction between two bodies, which is a function of their masses and the distance between them, and is described by Newton´s law of universal gravitation. An effect of gravity is observed when the fruit from a tree falls to the ground. The gravity field at each location on Earth consists of a global field which is superimposed by a local anomaly field. In a gravity survey, measurements are made of the local gravity field differences due to density variations in the subsurface. The effects of smallscale masses are very small compared with the effects of the global part of the Earth’s gravity field (often on the order of 1 part in 106 to 107).
Fig. 4.2-1 a-c: Principle of a gravity measurement: (a) the model shows a geological structure with a density U1 embedded in material with a higher density U2, (b) the spring with a small mass at the end of it changes length with changes in the gravity field, (c) the measurement results are plotted to document the gravity anomaly 'g(x)
186
4.2 Gravity Methods
Highly sensitive gravimeters are necessary for measuring such variations in gravitational attraction accurately. Special data processing and interpretation techniques (Section 4.2.6) are used to interpret the shape and amplitude of the anomalies in terms of subsurface geological or anthropogenic structures. Gravity measurements can be performed on land, at sea, and in the air. For environmental problems, land measurements are generally made. A necessary condition for the application of this method is the existence of density contrasts. The schematic in Fig. 4.2-1 shows a structure with a density U1 embedded in material with a higher density U2. This could be a channel in a boulder clay layer filled with sand or gravel or a pit filled with waste. Because of the negative density contrast (U1 - U2 < 0), the resulting gravity anomaly 'g is negative too. Considering a gravimeter as basically a mass on a spring, the amplitude of the gravity anomaly is a function of the expansion or contraction of the spring, the geological situation in Fig. 4.2-1a will cause the length of the spring to decrease above the anomalous structure.
4.2.2 Applications x Structural investigation of landfill, industrial or mining sites and their surroundings, x obtaining structure and thickness of unconsolidated sediments, x lithological subdivision of the subsurface, particularly in areas of unconsolidated rocks, x detection of lithological and structural changes as well as fractures in consolidated rock, x detection of cavities, x locating concealed waste dumps, x estimation of the thickness and/or mean density of waste deposits, x detection of density inhomogeneities inside a waste dump (a rare application), and x gravity data are also used to provide constraints in the interpretation of seismic data.
4.2.3 Fundamentals Each point P on, above and below the Earth’s surface is affected by the gravitational (or gravity) field g(P). This is a natural potential field like, for example, the magnetic field. The gravity field is measured in units of
Environmental Geology, 4 Geophysics
187
acceleration [m s-2] 1 . This field is mainly caused by the attraction between masses of the Earth and an arbitrary mass at point P with the coordinates x, y, z. The gravitational force F between two point masses m and m’ with the distance r between them is: F ( P)
G
mm' r2
(4.2.1)
where the gravitational constant G = 6.672 10-11 m3 kg-1 s-2. The effect of the gravitational field g of m on m’ is derived from Equation (4.2.1): g E ( P)
G
m . r2
(4.2.2)
This effect is called gravitational acceleration. The mass m of a body is given by the product of its density U and volume V. In addition to the gravitational acceleration of the Earth gE (P), point P is affected by centrifugal acceleration gC (P) and by the gravitational attractions of mainly the moon and the sun gT (P), the variations of which are called tides. These tides vary with respect to place and time. The centrifugal acceleration gC (P) is due to the rotation of the Earth and depends on the latitude M of the point P. The gravitational field at point P is: g(P) = gE (P) + gC (P) + gT (P).
(4.2.3)
This formula describes the global gravity field. The calculation of the individual components of the global field is described in Section 4.2.6. The anomalous gravity field 'g caused by density inhomogeneities of geological or anthropogenic structures is superimposed on the global field. Equation (4.2.2) can be used to calculate 'g if the mass m is replaced by the anomalous mass 'm = 'U V, where 'U is the density contrast between the inhomogeneity and the surrounding material, V is the volume of the inhomogeneity. Consequently, the gravity value g(P) at point P is calculated from the theoretical gravity field value J0, the tidal effect and the anomalous field: g(P) = J0 (P) + gT (P) + 'g(P).
(4.2.4)
For geophysical surveys, only gravity anomalies 'g(P) related to density inhomogeneities in the subsurface are of interest.
1 In the SI system, the unit of gravity acceleration is 1 µm s-2 (= 10-6 m s-2), called a gravity unit (g.u.), but in practice the older unit 1 mGal (= 1 milliGal = 103 µGal = 10 µm s-2 = 10-5 m s-2) is often used. Thus 0.1 mGal = 1 g.u.
188
4.2 Gravity Methods
Petrophysical basis Gravity measurements are only useful if the density of the target is significantly different from the density of the surrounding material. Therefore, it is important to have an idea of the typical density values of the materials in the area to be investigated. The density values of selected materials, including typical waste, are given in Table 4.2-1. Table 4.2-1: Typical densities of selected sedimentary, metamorphic and igneous rocks, and waste, = mean value Sediments fine sand, dry fine sand, wet medium sand gravel, dry gravel, wet silt clay, dry clay, wet limestone sandstone rock salt Metamorphic and igneous rocks granite granodiorite basalt gneiss Waste organic high ash content domestic, mixed high rubble content
Density in g cm-3 (103 kg m-3) 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8
1.5 1.8 1.7 1.7 2.0 1.7 1.7 1.9 2.55 2.35 2.2
2.65 2.7 3.1 2.7
1.25 1.3 1.5 1.8
The density of waste material is almost always less than that of the surrounding rock, especially consolidated rock. Thus, there is a negative gravity anomaly, sometimes called a gravity minimum or gravity low, above a landfill. If no density data are available in the investigation area, a few cone penetration tests (in unconsolidated material) or geophysical borehole logs (e.g., gamma-gamma logs, sometimes called density logs) can help to obtain basic data.
189
Environmental Geology, 4 Geophysics
Gravity anomalies of simple structures Calculation of the gravity anomalies of simple structures is useful in different phases of a gravity survey. In the project planning phase this can help to decide whether gravity data would be useful and which anomalies are to be expected. Based on this knowledge, the parameters and equipment for the survey can be selected. In the interpretation phase, the measured gravity anomalies are first interpreted qualitatively, but the main aim of the interpretation is to obtain information about the size and depth of the structures of interest. To obtain such a quantitative interpretation, it is necessary to calculate the theoretical gravity anomaly that would be caused by a geological structure and to compare the result with the measured data. Geological structures usually have a complicated geometry and their effect on the gravity field cannot be sufficiently described by Equation (4.2.2). Therefore, Equation (4.2.2) is replaced by the integral expression (4.2.5). For all points P(x, y, z) on the Earth’s surface (z = 0) the gravity anomaly caused by an anomalous body with a density difference 'U to the host material is: 'g ( x, y, 0)
zc
G 'U ³ 3 dx ' dy ' dz ' V r
(4.2.5)
where r2 = (x – x’)2 + (y – y’)2 + z’2; x, y, z are the coordinates of the point of measurement and x’, y’, z’ are the source points of the body. The gravity anomalies of several structures (forward modeling) can be calculated by using different integration paths in Equation (4.2.5) corresponding to the boundaries of the source body. In general, the modeling can be done for 2-D or 3-D structures: Two-dimensional structures:
The length L of the source body (y’-direction) is quasi-infinite, i.e., much larger than its width W in the x’ and z’-direction (L t 4W), permitting integration only across its crosssection F in the x, z-plane.
Three-dimensional structures:
The source body is finite in the x’, y’ and z’-directions; integration is over the volume V of the body as shown in Equation (4.2.5).
Simple examples of both types of structures are a horizontal cylinder and a sphere. These shapes can easily be used for a rough calculation of the gravity anomalies of similar geological structures. Assuming a horizontal cylinder and a sphere with a radius R and the center of the structure at a depth d (d > R) on the z-axis and the density contrast 'U with respect to the host material, then
190
4.2 Gravity Methods
the gravity anomalies of these bodies at the Earth’s surface z = 0 can be calculated as follows (LINDNER & SCHEIBE, 1978): 'g ( x) cylinder
2S R 2 G 'U
d x2 d 2
(4.2.6)
and 'g ( x) sphere
d 4S 3 R G 'U 2 3 ( x d 2 ) 3/ 2
.
(4.2.7)
According to (4.2.6) and (4.2.7) the maximum or minimum of the gravity anomaly is at x = 0, above the center of a homogeneous body. Figure 4.2-2 shows the anomalies of horizontal cylinders of same size at different depths. As can be seen in Fig. 4.2-2 the anomaly amplitude decreases with increasing depth. On the other hand, the width of the anomaly increases with increasing depth.
Fig. 4.2-2: Gravity anomaly of a horizontal cylinder calculated for different depths d, density contrast between the cylinder and its surroundings 'U = -1.0 u103 kg m-3, R = 2 m, center of the cylinder at x = 0 m, A – detection limit of the gravity measurements as defined by the precision
Environmental Geology, 4 Geophysics
191
A collection of formulas for other simple models is given by TELFORD et al. (1990), and LINDNER & SCHEIBE (1978). Such simple models can be used only for a rough interpretation of complicated geological structures. For more than a rough interpretation, software has to be used for two-dimensional and/or three-dimensional modeling (GÖTZE & LAHMEYER, 1988; BOULANGER & CHOTEAU, 2001). The latter is rather expensive and requires considerable experience in gravity modeling (Section 4.2.6). Ambiguity of the results The gravity field is a potential field. Therefore, it is not possible to obtain unequivocal information about depth, extent and/or density from gravity measurements alone (owing to the principle of equivalence). This ambiguity can be reduced or even be removed by inclusion of the following additional information when the gravity data is interpreted: -
all available geological information about the survey area,
-
reliable density data of the materials expected,
-
inclusion of the results of other geophysical methods (e.g., seismic and geoelectric methods).
Thus, the most useful results of gravity methods are obtained within the scope of an integrated geophysical survey (SILVA et al., 2002).
4.2.4 Instruments The gravity field is measured by a gravimeter. Gravimeters used in a gravity survey can only measure the difference in gravity between two observation points, not the absolute values. Therefore, relative gravity values are measured in a survey, i.e., the measured data at single stations is related to a reference station with a known gravity value. The results of the local survey are linked to a larger gravity network in this way. If there is no reference station available in the investigation area, one station can be defined as reference station with an arbitrarily chosen value. The absolute values are generally not important for environmental surveys. All instruments used in field geophysics employ some type of static springmass system, i.e., the gravitational force on a mass in the gravimeter is balanced by a spring, and the change in length of the spring is measured. Most gravimeters are of the astatic type to increase the precision. In this type of gravimeter, the amount necessary to balance the mass to zero is determined. There are two main spring systems: The first is based on a concept developed by S. P. Worden in 1947 and uses quartz springs (e.g., the Sodin and Sharpe gravimeters); the second system, developed by LaCoste & Romberg, uses
192
4.2 Gravity Methods
metal springs. Most instruments house their systems in a vacuum and have thermostats to control temperature drift, keeping the drift to a minimum and increasing the precision (CHAPIN et al., 1999; CARBONE & RYMER, 1999). Gravimeters with manual adjustment of the length of the spring and automated gravimeters with an electronic feedback system are available. The instruments with manual compensation require experienced operators and do not allow digital data storage. But, these instruments are less expensive than automated gravimeters, which use electrostatic systems to adjust the length of the spring. The measured gravity values, time, instrument tilt, drift, etc. are stored in RAM and can be transferred to a PC. State-of-the-art automated gravimeters (e.g., CG-5, Scintrex Ltd.; Graviton EG, LaCoste & Romberg) have the following specifications: sensitivity
0.001 to 0.005 mGal
precision in the field
r 0.005 to 0.01 mGal
dynamic range without resetting
8000 mGal
measuring time per station
2 - 3 minutes (in areas with low seismic noise)
temperature range
-40° to + 45° or 0 to 55°C
Instruments with manual adjustment (e.g., Sodin 410, W. Sodin Ltd.; LCR G or D from LaCoste & Romberg) have a precision of 0.005 to 0.010 mGal. This is sufficient for most survey purposes, but for the detection of cavities, instruments with a higher precision should preferably be used.
4.2.5 Survey Practice Planning of a survey Planning of a survey is an important step in exploration because the selection of appropriate survey parameters strongly influences the quality of the results. The following survey parameters are to be determined: 1. Size of the survey area: The survey area should be larger than the target (e.g., a covered dump site) in order to record the entire gravity anomaly, which is larger than the target itself, and to allow the definition of the background, or regional field. 2. Selection of profile and station spacing: If possible, the measurements should be done at almost equidistant grid points. If the measurements can only be made along profiles, the distance between the profiles should not be more than 2 - 4times the distance between stations along a profile. The station spacing depends on the information needed and on the depth and
Environmental Geology, 4 Geophysics
193
size of the expected structure. For the detection of cavities or local density inhomogeneities the expected anomaly should be covered at least by five stations. The magnitude and areal extent of the expected anomalies can be calculated using the simple models mentioned in Section 4.2.3 or by sophisticated model calculations. 3. Selection of a location for a reference station: Because the gravity value depends on the elevation, the gravity value and station elevation must be measured for each field station. At least one station must serve as reference for both measurements. This station should be in a seismically quiet area and it should not be possible for its value to change during the survey. In large survey areas, it is useful to establish several reference stations. Determination of coordinates and elevation Determination of the coordinates and elevation of a gravity station is an indispensible part of a gravity survey. The values for all stations are to be related to the reference station. The measurements can be done by leveling, by electro-optical tachymetry, or by GPS (Global Positioning System). The advantage of the latter two is that the coordinates of the station can be determined simultaneously with the elevation. In the case of levelling, the position of the stations has to be determined using a tape measure. For small areas and small station spacings (d 25 m), a regular grid can be advantageous. For larger station spacings, the locations of the stations are drawn on a good quality topographic map and the coordinates are determined by digitization. The selection of the appropriate and most efficient method depends on the size and the topographic conditions of the survey area. In any case, the elevation of a station should be determined with a precision better than r 0.05 m (in order to obtain a precision of the gravity value better than 0.01 mGal). See also Chapter 2.5. Gravity survey All gravity measurements at the field stations are related to a single reference station or to one of the reference stations of the base net (if established). The tie to the reference station(s) can be obtained by either single-loop measurement: all field stations are measured once, or double-loop measurement: all field stations are measured twice. The double-loop method requires almost double the effort of the single-loop method but it is useful if very high precision is desired (e.g., when underground cavities are being searched for).
194
4.2 Gravity Methods 0.20
0.20
0° 0.15
0.15
0.10
Earth tide correction in mGal
0.10
40°N 0.05
0.05
40°S 0.00
0.00
-0.05
-0.05
80°N -0.10
-0.10
-0.15
-0.15
-0.20
-0.20 0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Time in h
Fig. 4.2-3: Earth tide correction for different latitudes along 15° E on 4th October 2002; the tidal curve is superimposed by the drift curve of the instrument (the daily drift of a good quality instrument should be less than 0.02 mGal)
The measurement at the reference station has to be done at least at the beginning and at the end of each loop. As the measured value varies with time due to the Earth tide (Fig. 4.2-3) and the drift of the instrument, both of these parameters have to be determined. Normally, only a theoretical value is calculated for the Earth tide correction. Instrument drift must be determind by repeated measurement at the base station within a certain time interval. This interval depends on the quality and specifications of the gravimeter and can be between 30 and 240 minutes. If the Earth tide correction cannot be calculated, a tie-back to a base station on a hourly basis can be done to record and correct the superimposed influence of the Earth tide and the instrument drift. For quality control, some stations (5 - 10 %) of previous loops should be measured repeatedly in each loop.
Environmental Geology, 4 Geophysics
195
4.2.6 Processing and Interpretation of the Measured Data Corrections The gravity value is influenced by the elevation and the coordinates of the gravity station, the time of the measurement, and the surrounding morphology. Therefore, it is necessary to carry out several corrections. The aim of the corrections is to make the measurement at a single station comparable to the results at the other stations and to remove from the gravity value all known influences that are not due to the investigated structure (LEFEHR, 1991). The repeated measurements at the reference station in a loop are normally different from the first measurement in the loop. This is due to the effect of tides and instrument drift. These effects have to be eliminated using the internal software of the gravimeter or PC software. When the measurement at the reference station is repeated at least every 45 - 60 minutes, the differences from the first measurement are due to instrument drift and tidal changes; only when the measurements are repeated can the correction of both effects be carried out in one step. The result of this data processing is the drift and tidal corrected gravity value 'g. The resulting value due to unknown subsurface structures is called the Bouguer anomaly 'g0” and it is calculated as follows: 'g0” = 'g - J0 + GgFA - GgBoug + GgTop
(4.2.8)
The main corrections are described in Table 4.2-2. The interpretation is based on the Bouguer anomaly calculated using Equation (4.2.8). The data can be shown along profiles as a measured curve or plotted as a contour map. In most cases, the data have to be interpolated on a regular grid before plotting. A colored contour map improves the readability of the gravity map.
196
4.2 Gravity Methods
Table 4.2-2: The main gravity corrections Symbol
J0
GgFA
Name
Description
= Je (1 + E sin² M - E1 sin² 2M) in mGal where normal gravity (GRS67) (GRS80) 2 (theoretical gravity Je = 978031.8 mGal Je = 978032.7 mGal field) E = 0.005 3024 E = 0.005 3024 E1 = 0.000 0059 E1 = 0.000 0058 free air correction
= 0.3086 h
in mGal
in mGal GgBoug Bouguer correction 2S G = 0.04192 U h As the Bouguer correction assumes a flat, horizontal GgTop topographic
(terrain) correction area around the station, in hilly or mountainous areas it is necessary to correct for the effect of the masses in the surroundings that are not at the same elevation as the station. The topographic relief in the immediate vicinity of the station or survey area may require special surveying. This correction is calculated using special software. A template or a zone chart can also be used. h - station elevation in m U - density in g cm-3 or in 103 kg m-3 M - latitude of the gravity station G = 6.672 u 10-11 m3 kg-1 s-2 - gravitational constant
Interpretation Separation of the regional trend from the residual gravity anomaly: A Bouguer anomaly contour map contains all the gravity effects of both deep sources (the regional part of the gravity field) and shallow sources (the local or residual part). For a targeted interpretation, it is useful to separate the regional and the local parts of the field, in particular to make the anomalies caused by shallow sources more easily recognized. In general, the regional trend (field) is determined by various methods (e.g., graphical and smoothing techniques, gridding methods, filtering, HINZE (1988); TELFORD et al. (1990), or by modeling known structures). The regional field is subjectively choosen and it represents the gravity field that would have been measured if the gravity expression of the anomalous structure being investigated was not present. This regional field is subtracted from the Bouguer gravity and the residual field comprises mainly the local anomalies. The quality of the field separation 2 The Geodetic Reference System 1980 (GRS80) was adopted by the International
Association of Geodesy (IAG) during the General Assembly 1979 as reference system for size, shape, and gravity field of the Earth for geodetic, geophysical, astronomical and hydrographic applications. It replaced the Geodetic Reference System 1967 (GRS67). For principal parameters of GRS80 see http://dgfi2.dgfi.badw-muenchen.de/geodis/REFS/grs80.html.
Environmental Geology, 4 Geophysics
197
strongly influences the interpretation of the data, especially if the residual field is used as reference field for modeling the geological structures. Maps depicting regional or residual fields are both used for qualitative interpretation (PAWLOWSKI, 1995). Local field maps are important for the identification of tectonic structures in a survey area. Modeling: The main aim of interpretation is to provide quantitative information about the geology below the surface, i.e., depth, extent and petrophysical properties of structures (ABDERRAHMAN et al., 2001). Gravity modeling is necessary to obtain such information. Depending on the geology, 2-D, 2.5-D, 2.75-D or 3-D modeling is carried out. 2-D modeling is carried out along a profile if the structure (perpendicular to the profile direction) is much larger than it is wide. Each such structure along the profile is approximated by polygonal cross-sections. The calculated field is compared with the measured field and the model is modified until the calculated data sufficiently fits the measured data. 2.5-D or 2.75-D modeling will increase the accuracy of the modeling if the length of the structure is not quasi-infinite, i.e., less than four times the width. The most accurate modeling of the subsurface geology is done with 3-D modeling. Programs for inverting surface gravity data to derive a 3-D distribution of density contrast are also available. Geological units are constructed from triangular or rectangular elements and thus, very complex structures can be included in the model (GÖTZE & LAHMEYER, 1988; LI & OLDENBURG, 1998). The disadvantage of this type of modeling is that it requires special software, time-consuming model preparation and therefore, is often too expensive for small projects. Independent on the type of modeling, all available information (drill cores, borehole logs, density data, geological knowledge) has to be included in the modeling process to produce the most reliable model of the subsurface (BOULANGER & CHOTEAU, 2001; BARBOSA et al., 2002).
4.2.7 Quality Assurance The general aspects of quality assurance are given in the glossary. The following measures are important for quality assurance in a gravity survey: x Coordinates and elevation of the stations have to be determined with the required precision (Section 4.2.5). The errors in coordinates and elevation of observation stations have to be documented. x Documentation of the coordinates and elevations of all gravity reference stations and benchmarks. Reference stations should be permanently marked in the field.
198
4.2 Gravity Methods
x The gravimeters should be calibrated at least once a year at base stations with at least 80 mGal difference. The drift of the gravimeters and function tests should be documented. x To determine the error of the gravity measurements, repeated measurements should be made in different loops at 5 - 10 % of the gravity stations. x The quality of the data should be checked during the field work, permitting errors to be detected so that measurements can be repeated at these stations, if necessary. x The corrections are to be documented. In addition, information about datum planes and the density values used for reductions is necessary. x Uncertainties in the interpretation, in particular in the modeling, should be given.
4.2.8 Personnel, Equipment, Time Needed Personnel mobilization and demobilization measurements: position, elevation
gravity survey
Equipment
Time needed
depends on the distance to the survey area 1 surveyor 1 assistant
1 operator (1 assistant)
1 instrument for surveying ,
a = 5 - 10 m L = 200 N/d
1 4WD vehicle
a = 200 m L = 30 N/d
1 gravimeter,
a = 5 - 10 m L = 80 - 120 N/d
1 4WD vehicle
a = 200 m L = 30 N/d
1 PC, (e-mail for data transfer) data processing interpretation report
1 geophysicist (1 technical assistant)
1 PC with plotter, printer and software; (e-mail for data transfer)
3 - 4 days for each day in the field
a = station spacing, N = number of stations, d = 10-hour working day, L = N/d
199
Environmental Geology, 4 Geophysics
4.2.9 Examples Investigation of a concealed waste dump in an abandoned quarry This example describes the investigation of a concealed waste dump in a former sandstone quarry in southern Germany. The objective was to determine the location and size of the landfill. Survey parameters: size of the investigation area
12 000 m2
number of stations
353
repeated stations
10 %
station spacing
5 m, at the margin 10 m
duration of the survey
6 days
personnel
1 geophysical technician, 1 assistant
measured physical parameter
'g
equipment
Sodin 410,
corrections and reductions
drift correction free air correction normal gravity (relatively calculated) Bouguer reduction using a density of 2.0 103 kg m-3
The Bouguer anomaly map of the area under investigation is shown in Fig. 4.2-4a. The gravity low in the southeastern part of the area is assumed to coincide with the location of the former quarry, now filled with waste and covered with soil. The edge of the landfill can be identified clearly, both laterally and with depth. The gravity high east of the gravity low is caused be sandstone. To the north and west the thickness of the weathered material covering the sandstone increases. An estimate of the depth of the waste is given in Fig. 4.2-4b and c. For modeling, the residual field is calculated by subtracting an almost horizontal linear trend from the Bouguer anomalies. Depending on the density contrast assumed, the maximum depth of the former quarry is between 9 m and 20 m. As the density difference and the depth of the waste are not known, this ambiguity inherent in gravity data can only be removed by further investigation, e.g., by boreholes in the area of the gravity anomaly or by additional geophysical investigations such as seismics or geoelectrics. Other examples of gravity modeling of waste pits are given in Fig. 5.4-24.
200
4.2 Gravity Methods
Fig. 4.2-4: Gravity survey of a concealed waste dump in a former quarry, (a) Bouguer anomaly map, (b) comparison of the measured and calculated gravity anomalies of the profile A - B for the two models shown in (c)
201
Environmental Geology, 4 Geophysics
Investigation of structures in the area around an open pit mine in Sarawak, Malaysia A short time after the re-opening of an open-pit mine directly south of the town of Bau in Sarawak, there was some subsidence and sinkholes appeared. In order to establish a correlation between the sinkholes, subsidence, and geological structures such as faults and other types of contacts, staff of the Geological Survey of Malaysia, Sarawak (GSMS), assisted by German consultants, carried out a gravity survey of the entire municipal area of Bau. Additionally some seismic lines were surveyed. The structure in this area is characterized by nearly vertical contacts between limestone and unconsolidated materials (e.g., mudstone, tailings). The limestone is karstified and it was thought that collapsing solution cavities were responsible for the sinkholes. Survey parameters: size of the investigation area
| 1.5 km2
number of stations
658
repeated stations
46
station spacing
40 - 50 m
duration of the survey
13 days
personnel
1 geophysical technician, 1 assistant
measured physical parameter
'g
equipment
LCR Model G
corrections
drift correction free air correction normal gravity (formula: GRS67) Bouguer correction using a density of 2.0 103 kg m-3
precision of corrected data
0.018 mGal
The main results of the survey are shown in Fig. 4.2-5. This residual map was prepared by applying a high-pass wavelength filter with a cutoff wavelength of 1 km to the Bouguer anomaly data. The gravity highs (yellow to red) coincide with areas where the overburden above the limestone has a very small thickness. Outside the areas of these gravity highs, the soil is generally thicker. Several local gravity lows (e.g., in the NW, N and SE part of the survey area) coincide with areas in which shale has been mapped. In these gravity-low areas the shale can be expected to be very thick.
202
4.2 Gravity Methods
Fig. 4.2-5: Gravity survey of structures in the area around an open-pit mine in Sarawak, Malaysia, contour map of the residual gravity field obtained using a high-pass wavelength filter with a cutoff wavelength of 1 km
203
Environmental Geology, 4 Geophysics
Besides the information about variation in overburden thickness, the gravity data showed the locations of several faults (some of them as extensions of known faults) and provided a detailed outline of the structures in the survey area. These structures were later confirmed by a seismic survey. The main geological structure, starting from the northern edge of the pit, trends NNE-SSW. Most of the larger sinkholes in the town are along this structure, in particular to the crossing of two faults (Tai Parit fault and Bukit Young fault). This zone west of the strongest gravity high coincides with a zone where the depth to the limestone abruptly increases (high gradients in the gravity anomaly) and whose steep edges are in a direct contact with unconsolidated material. On the basis of the geophysical results, it is now assumed that groundwater moves along the faults and in the karst system. At the contact of the limestone with the unconsolidated material, the latter is removed by water producing cavities, which finally collapse. Another example of a structure map based on residual gravity is given in Fig. 5.4-21. Parameters and units Symbol
Units
gravitational acceleration
g
µm s-2, mGal, gravity units (g.u.)
density
U
kg m-3 (g cm-3) 1 g cm-3 { 103 kg m-3
gravitational constant
G
6.672 u 10-11 m3 kg-1s-2
mass
m
kg
References and further reading ABDELRAHMAN, E-S., M., EL-ARABY, T. M. EL-ARABY, H. M. & ABO-EZZ, E. R. (2001): A new method for shape and depth determinations from gravity data. Geophysics, 66, 1774-1780. ARAFIN, S. (2004): Relative Bouguer anomaly. The Leading Edge, 850-851. BARBOSA, V. C. F., SILVA, J. B. C. & MEDEIROS, W. E. (2002): Practical applications of uniqueness theorems in gravimetry: Part II – pragmatic incorporation of concrete geologic information. Geophysics, 67, 795-809. BOULANGER, O. & CHOTEAU, M. (2001): Constraints in 3D gravity inversion. Geophys. Prosp., 49, 265-280. CARBONE, D. & RYMER, H. (1999): Calibration shifts in a LaCoste-and-Romberg gravimeter: comparison with a Scintrex CG-3M. Geophys. Prosp., 47, 73-83. CHAPIN, D. A., CRAWFORD, M. F. & BAUMEISTER, M. (1999): A side-by-side test of four land gravity meters. Geophysics, 64, 765-775.
204
4.2 Gravity Methods
GIBSON, R. I. & MILLEGAN, P. S. (1998): Geologic applications of gravity and magnetics: Case histories. SEG, Tulsa. GÖTZE, H.-J. & LAHMEYER, B. (1988): Application of three-dimensional interactive modelling in gravity and magnetics. Geophysics, 53, 1096-1108. HINZE, W. J. (1988): Gravity and magnetic methods applied to engineering and environmental problems. Proceedings of the “Symposium on the Application to Engineering and Environmental Problems”, March 28 - 31, Golden, Colorado, 1-107. HINZE, W. J. (2003): Bouguer reduction density, why 2.67?. Geophysics, 68, 1559-1560. LEFEHR, T. R. (1991): Standardization in gravity reduction. Geophysics, 56, 1170-1178. LI, Y. & OLDENBURG, D. W. (1998): 3-D inversion of gravity data. Geophysics, 63, 109-119. LI, X. (2001): Vertical resolution: Gravity versus vertical gravity gradient. The Leading Edge, 901-904. LITINSKY, V. A. (1989): Concept of effective density: key to gravity depth determinations from sedimentary basins. Geophysics, 54, 1474-1482. LINDNER, H. & SCHEIBE, R. (1978): Die Berechnung von Gg- und GT- Anomalien für regelmäßige homogene Störkörper. Gerlands Beitr. Geophys., Leipzig, 87, 29-45. MILITZER, H. & WEBER, F. (1984): Angewandte Geophysik. 1. Springer, Wien. PARASNIS, D. S. (1997): Principles of Applied Geophysics. Chapman & Hall, London. PAWLOWSKI, R. S. (1995): Preferential continuation for potential-field anomaly enhancement. Geophysics, 60, 390-398. SHARMA, B. & GELDART, L. P. (1968): Analysis of gravity anomalies using Fourier transforms. Geophys. Prosp., 16, 77-93. SILVA, J. B. C., MEDEIROS, W. E. & BARBOSA, V. C. F. (2002): Practical applications of uniqueness theorems in gravimetry: Part I - constructing sound interpretation methods. Geophysics, 67, 788-794. Simon Fraser University, Vancouver (2003): Earth-sciences courses. www.sfu.ca/earth-sciences/courses/new207. TALWANI, M. & EWING, M. (1960): Rapid computation of gravitational attraction of threedimensional bodies of arbitrary shape. Geophysics, 25, 203-225. TELFORD, W. M., GELDART, L. P. & SHERIFF, R. E. (1990): Applied geophysics (2nd edn.). Cambridge University Press, Cambridge, New York, Port Chester, Melbourne, Sydney.
Environmental Geology, 4 Geophysics
205
4.3 Direct Current Resistivity Methods KNUT SEIDEL & GERHARD LANGE
4.3.1 Principle of the Methods Direct current (dc) resistivity methods use artificial sources of current to produce an electrical potential field in the ground. In almost all resistivity methods, a current is introduced into the ground through point electrodes (C1, C2) and the potential field is measured using two other electrodes (the potential electrodes P1 and P2), as shown in Fig. 4.3-1. The source current can be direct current or low-frequency (0.1 - 30 Hz) alternating current. The aim of generating and measuring the electrical potential field is to determine the spatial resistivity distribution (or its reciprocal - conductivity) in the ground. As the potential between P1 and P2, the current introduced through C1 and C2, and the electrode configuration are known, the resistivity of the ground can be determined; this is referred to as the “apparent resistivity”.
Fig. 4.3-1: Principle of resistivity measurement with a four-electrodes array
206
4.3 Direct Current Resistivity Methods
Resistivity measurements may be made at the Earth’s surface, between boreholes or between a single borehole and the surface. With special cables, measurements can be made underwater in lakes, rivers and coastal areas. The following basic modes of operation can be used: x profiling (mapping), x vertical electrical sounding (VES), x combined sounding and profiling (two-dimensional resistivity imaging), x three-dimensional resistivity survey (3-D resistivity imaging), and x electrical resistivity tomography (ERT). Profiling methods use fixed electrode spacings to detect lateral resistivity changes along a profile down to a more or less constant investigation depth, which is governed by the electrode spacing. The results are normally interpreted qualitatively. Contour maps or profile plots of the measured apparent resistivities allow delineation of lateral boundaries of geogenic structures and anthropogenic features (e.g., waste dumps and contamination plumes), as well as hydrogeological conditions. The main aim of sounding methods is to determine the vertical distribution of the resistivity in the ground. Several soundings at regular spacings along profiles and/or randomly in the area under investigation will also provide information about the lateral extent of structures. Soundings may be made for investigation depths up to several hundred meters. The measured data may be interpreted both qualitatively and quantitatively. The latter will provide resistivity models whose layer boundaries are boundaries of geoelectrical layers but not necessarily of lithological layers. For a better correlation to the geology, geoelectrical data should be correlated with borehole logs or the results of other geophysical methods, such as reflection and refraction seismic sections. Sounding and profiling can be combined in a single process (2-D resistivity imaging) to investigate complicated geological structures with strong lateral resistivity changes. This combination provides detailed information both laterally and vertically along the profile and is the most frequently applied technique in environmental studies. 2-D inversion yields a two-dimensional distribution of resistivities in the ground. Three-dimensional (3-D) resistivity surveys and ERT measurements provide information about complex structures. In current practice they do not play an important role in geophysical site investigations, as they are still very time consuming and expensive. Some convincing examples in small survey areas show that new and effective 3-D techniques are being developed, including data acquisition and interpretation (DAHLIN et al., 2002).
Environmental Geology, 4 Geophysics
207
4.3.2 Applications x
Investigation of lithological underground structures,
x
estimation of depth, thickness and properties of aquifers and aquicludes,
x
determination of the thickness of the weathered zone covering unweathered rock,
x
detection of fractures and faults in crystalline rock,
x
mapping of preferential pathways of groundwater flow,
x
localization and delineation of the horizontal extent of dumped materials,
x
estimation of depth and thickness of landfills,
x
detection of inhomogeneities within a waste dump,
x
mapping contamination plumes,
x
monitoring of temporal changes in subsurface electrical properties,
x
detection of underground cavities,
x
classification of cohesive and non-cohesive material in dikes, levees, and dams.
4.3.3 Fundamentals Physical basics A point electrode introducing an electrical current I will generate a potential Vr at a distance r from the source. If both source and measuring points are at the surface of a homogeneous half-space with resistivity U, this potential is given by:
U,
(4.3.1) 2S r In the case of a four-electrode array (Fig. 4.3-1) consisting of two current electrodes (C1, C2) that introduce a current ± I, the potential difference 'V between the potential electrodes P1 and P2 can be calculated as follows: Vr
ª1 §1 1 1 1 ·º 'V U , « ¨ ¸» , r2 r3 r4 ¹ ¼» ¬« 2S © r1
where r1 = C1P1, r2 = C1P2, r3 = C2P1, and r4 = C2P2.
(4.3.2)
208
4.3 Direct Current Resistivity Methods
Replacing the factor in square brackets by 1/K, we obtain the resistivity of the homogeneous half-space as follows:
U .
'V . I
(4.3.3)
The parameter K, the configuration factor or geometric factor, can be easily calculated for all practical configurations. Table 4.3-1 gives K parameters for arrays most commonly used in field surveys. For inhomogeneous conditions, Equation (4.3.3) gives the resistivity of an equivalent homogeneous halfspace. For this value the term apparent resistivity Ua is introduced, which is normally assigned to the center of the electrode array. In a given fourelectrode array, the current and potential electrodes are interchangeable. This concept (the principle of reciprocity) can be used in multi-electrode systems to improve the signal-to-noise ratio. There are many different types of electrode arrays (also called spreads or configurations) although in practice only a few of them are used. The most common are compiled in Table 4.3-1. Each array has its advantages and disadvantages with regard to depth of investigation, resolution of horizontal and vertical structures, sensitivity to lateral changes in resistivity (lateral effects) and inhomogeneity, to depth of targets, to dip and topography, etc. Depending on the array characteristics and the objective, the appropriate array has to be selected for each survey. Field logistical qualities are also essential for a survey. A comprehensive assessment of electrode arrays is given by WARD (1990). The type of electrode array selected for a resistivity survey plays a role with regard to resolution and depth of investigation. For resistivity soundings (VES), the vertical resolution refers to how thin a layer can be detected in a layer sequence. In recent years 2-D resistivity measurements with a multielectrode system have become increasingly important, especially for environmental applications (BARKER, 1981; DAHLIN, 1996). The vertical and horizontal resolution of dipping structures and lateral resistivity distribution is of interest in the case of a 2-D survey. At the same time as the new measuring techniques, 2-D inversion schemes have been developed using the increasing capacity of personal computers (DEY & MORRISON, 1979a; BARKER, 1992). A sensitivity matrix is an important part of the inversion algorithms. For a given electrode array, this matrix is used to evaluate the contribution of the spatial elements of the subsurface model to the measured apparent resistivity. Sensitivity matrices are very helpful for the understanding of measured data and, especially for the planning a survey, as they allow conclusions to be drawn about the resolution capabilities and investigation depths of different electrode arrays. For the case of a homogeneous half-space, some examples of sensitivity matrices for commonly applied electrode arrays are given in Fig. 4.3-2. Each of these examples shows that there are areas of positive as well as of negative
Environmental Geology, 4 Geophysics
209
sensitivity. In an area of negative sensitivity, there will be a decrease in the measured apparent resistivity when the structure in that area has a higher resistivity than the surrounding material. Negative sensitivities always occur in the space between a current and a potential electrode, whereas the sensitivity is always positive between two current electrodes and between two potential electrodes (Fig. 4.3-2). The spatial sensitivity generally decreases with increasing distance from the electrodes (FRIEDEL, 2000). The plots in the left side of Fig. 4.3-2 show sensitivities for small electrode spacings, whereas those on the right side show sensitivities for a larger electrode spacing. The different arrays can be assessed on the basis of these sensitivity plots. It can be seen, for example, in plots (c) and (d) that the sharp, almost vertical boundary between positive and negative sensitivities is the reason why pole-dipole arrays have a very good lateral resolution. The fairly high resolution of dipole-dipole measurements in the case of small targets results from a structured sensitivity distribution (Fig. 4.3-2e and f). But it can also be seen that inhomogeneity near the electrodes strongly influences the measured apparent resistivity. Experience shows that the results of Schlumberger arrays are mainly determined by the ground conditions below the potential electrodes, as Fig. 4.3-2g and h confirm. Finally, the fairly smooth sensitivity distributions shown in plots (i) and (j) explain the comparatively low lateral resolution of a Wenner array. But this array is advantageous if the investigation area is subject to electromagnetic noise. Two examples of dipole-dipole-measurements with electrodes in boreholes are given in Figs. 4.3-2k and l. Figure 4.3-2m shows the sensitivity of a Wenner array in the x-y-plane at the surface. It can be seen that differences in the rock to the side of the profile will also influence the measured data. This is very important to know if misinterpretation is to be avoided in areas with a complicated 3-D geology. In general, the geology of interest has a 3-D resistivity distribution. Normally, this would require a 3-D resistivity survey. But this is, as already mentioned, very time consuming and expensive at present. In many practical cases the problem can be reduced to a 2-D or even a 1-D case, but it must always be kept in mind that the measurements can be influenced by objects outside the arrays (SPITZER, 1995). In the 2-D case it is assumed that the resistivity of the ground varies only in the vertical and one horizontal direction. There is no resistivity variation in the second horizontal direction (strike direction). As a consequence, survey profiles should run perpendicular to the strike of such structures.
210
4.3 Direct Current Resistivity Methods
Table 4.3-1: Some electrode arrays for dc resistivity measurements
Environmental Geology, 4 Geophysics
211
Fig. 4.3-2: Sensitivity of several electrode arrays, examples for the 2-D-sensitivity distribution in a homogeneous half-space: (a,b) pole-pole, (c,d) pole-dipole, (e,f) dipoledipole, (g,h) Schlumberger array, (i,j) Wenner array, (k,l) dipole-dipole with electrodes in boreholes, (m) Wenner array but x-y-plane of the sensitivity. All plots are normalized with respect to the maximum of each matrix. (courtesy of S. FRIEDEL, University of Leipzig, for details on computation, see FRIEDEL, 2000).
212
4.3 Direct Current Resistivity Methods
In the 1-D case, it is assumed that the ground corresponds to a horizontally layered model. In practice, this requirement is fulfilled if the dip of the layers is less than 10°. The 1-D case is very often assumed for vertical electrical soundings. In any case, the possibility of 2-D or 3-D effects must be taken intro consideration when the sounding survey is planned. The limits of 1-D inversion modelling of a 2-D situation are discussed by SCHULZ & TEZKAN (1988) and BEARD & MORGAN (1991). Ambiguity of results The dc resistivity method, being a potential field method, exhibits considerable inherent ambiguity. In the case of resistivity sounding, the ambiguity is related to layer thickness, layer resistivity and governed by the principles of equivalence and suppression. The principle of equivalence in the 1-D case means that it is impossible to arrive at a unique solution for the layer parameters (thickness and resistivity). For conductive layers, only the thickness/resistivity ratio can be determined (S-equivalence), whereas for highly resistive layers only the product of thickness times resistivity can be determined (T-equivalence). This means that when resistivity and thickness of a layer vary within certain limits, their product remaining constant, no differences can be seen in the sounding curve. Thickness and resistivity are coupled in both cases of equivalence and cannot independently determined. Because this principle results in different equivalent layer models which all fit the sounding curve within a selected fitting error range, it is important to correlate resistivity sounding data with data from boreholes in the area of investigation. The principle of suppression (hidden layer problem) applies when an intercalated layer has a resistivity intermediate between the resistivities of the layers above and below. In this case, the layer has an insignificant effect on the sounding curve unless it is very thick. A rule of thumb says that a layer can be detected if its thickness is greater than its depth and its resistivity differs from the cover layer. By modeling of sounding data, a series of thin layers will mostly be represented by only one layer and a mean resistivity. There are still problems with the determination of the confidence intervals and parameter limits of models for the 2-D and 3-D cases. In 2-D inversion, ambiguity is influenced by several factors: the structure of the grid used to approximate the geological structures, limited data density, and errors in the data. Not to be neglected is the sensitivity distribution of the electrode configuration used (SPITZER & KÜMPEL, 1997; FRIEDEL, 2000; Fig. 4.3-2). 2-D equivalence of smooth and sharp boundary inversion in the investigation of simple structural models (e.g., vertical fault, graben, horst) using Wenner array is discussed by OLAYINKA & YARAMANCI (2002). Inversion using sharp boundaries (block inversion) indicates that - for good quality data with sufficiently high density - the range of 2-D equivalence is relatively narrow.
213
Environmental Geology, 4 Geophysics
For equivalent solutions the misfit between the observed data and the calculated data is very small in these cases. Models that are incorrect can be readily identified on the basis of very large data misfits.
Petrophysical basics Resistivity methods are useful only if the resistivity of the target differs significantly from the resistivity of the host material. For the successful planning of any geoelectrical survey and for the interpretation of the data, it is very important to know the resistivities of the materials in the study area. Resistivities of some selected materials, including typical domestic and industrial waste, are listed in Table 4.3-2. Table 4.3-2: Resistivities for geological and waste materials Material gravel sand silt loam clay (wet) clay (dry) peat, humus, sludge sandstone limestone schist igneous and metamorphic rock rock salt domestic and industrial waste natural water sea water (35‰ NaCl) saline water (brine)
Resistivity (in :m) minimum maximum 50 (water saturated) >104 (dry) 50 (water saturated) >104 (dry) 20 50 30 100 5 30 >1000 15 25 105 (compact) 100 (wet, jointed) >105 (compact) 50 (wet, jointed) >105 (compact) 106 (compact) 30 (wet) >106 (compact) 1000 (plastic) 10 300 0.25 10 M: and equipped with notch filters for 50 Hz and/or 60 Hz power-line rejection. The precision of the voltage readings of modern instruments is about 1 µV and that of the measured resistivities is about ±1 % within the operating temperature range from -20°C to 70°C. Most instruments offer signal enhancement by stacking and filtering as well as automatic self-potential tracking and suppression. Some receivers are able to also measure chargeability (Section 4.3.3). It is advantegeous if the equipment can be used for mapping/profiling and soundings, as well as 2-D and 3-D measurements. An additional switching unit is necessary for 2-D and 3-D measurements to activate the electrodes
216
4.3 Direct Current Resistivity Methods
individually. For some instruments a laptop is required for data acquisition, display and storage. The data acquisition software must be able to take into consideration the electrode configuration, electrode spacing and other survey parameters. Stainless steel electrodes seem to be best choice for the galvanic coupling of the electrodes with the ground, as they are durable and have fairly low selfpotentials, when placed in the ground. To obtain a sufficiently low contact resistance, the electrodes should be about 1 cm in diameter and at least 0.5 m long. For vertical electrical sounding with very large electrode spacing, a bundle of several metal rods can be used as current electrode in order to improve the galvanic coupling. In very dry ground the contact can be improved by pouring water around the electrodes. The type of cable used depends on the kind of measurement. Plastic- or rubber-insulated single-core cable with a cross-section of about 1.5 mm² is sufficient for profiling and sounding. Multi-core cables are used for 2-D/3-D resistivity measurements. Two types of operation are common. The first one uses “passive” electrodes connected to a switching box via a cable containing 20 or more cores. The other type of operation use “active” electrodes: A cable containing only a few cores – which are used to transmit current and communications, as well as measure voltage – is connected to an addressable box on each electrode. The addressable box is used to switch the electrode between active and passive modes. In practice, the first type is easier to handle, but the cable becomes very heavy for large basic electrode spacings (> 5 m). Crosstalk between cores can influence data acquisition in lowresistivity environments, and then it is better to use cables with separately shielded cores. The advantage of the second type is that the crosstalk is reduced because the cores all have a larger cross-section. Moreover, it is easier to increase the number of electrodes and to use basic electrode spacings of 10 m and more. For 3-D resistivity surveys multichannel instruments are essential to reduce data acquisition time to an acceptable level.
4.3.5 Survey Practice Planning a survey Before a survey is started, it is necessary to select the optimal kind of measurement (mapping/profiling, sounding or 2-D imaging). Such a decision is influenced by geology and topography in the study area, by the size of the area, and last but not least by economic considerations. An early discussion between the geophysicist and customer about the different geophysical, technical and economic parameters is recommended. For data processing and interpretation it is necessary to collect all available information about drilling results, the local geology and hydrogeology (particularly depth to the water
Environmental Geology, 4 Geophysics
217
table and groundwater flow direction) and potential sources of cultural noise (metal pipes, power lines, industrial objects). It is always very helpful if resistivity values for the materials of the geological structure to be investigated and information about possible contaminants in the soil and groundwater are available. In many cases, forward modeling of the expected situation can help to determine the most suitable method. It is essential to select the most appropriate electrode array for the problem at hand. Each array has advantages and disadvantages, which should be weighted against each other. Resolution and depth of investigation of different arrays have been investigated by several researchers (e.g., ROY & APPARAO, 1971; EDWARDS, 1977; BARKER, 1989; WARD, 1990). They all show that both the Schlumberger array and the Wenner array have very good vertical resolution, but due to their symmetric electrode geometry, a lower lateral resolution than asymmetric configurations, e.g., pole-dipole or dipole-dipole. Wenner arrays provide the best results under noisy conditions. Experience, confirmed by sensitivity matrix analysis (Fig. 4.3-2), suggests that the dipoledipole array has the best resolution with regard to the detection of single objects (e.g., cavities, sand and clay lenses). The investigation depth of commonly used arrays is, as a rule of thumb, in the range of L/6 to L/4, where L is the spacing between the two outer active electrodes. The dipole-dipole array, which offers the best depth of investigation (as long as the dipoles are small compared to the distance between them), has the disadvantage of a comparatively low signal-to-noise ratio (WARD, 1990). Therefore, to obtain good quality data, the electrodes must be adequately coupled to the ground and the equipment must have a high sensitivity. Further important points in the planning of a survey are size and location of the survey area. The survey area should be somewhat larger than the target (e.g., a concealed landfill) in order to properly determine its boundaries. The spacing of profiles and stations depends on the lateral resolution required to evaluate a certain geological situation - the more detailed the information needed, the smaller the interval between measurement stations. However, it must be remembered that the resolution decreases with increasing depth.
Influence of noise and topography Resistivity measurements in populated areas can be disturbed by various noise sources, such as grounded metal fences, underground metal pipes and cables, power lines, electric corrosion protection for pipelines or leakage currents generated by industrial facilities, streetcars and trains. Especially underground metal pipes can produce anomalies which influence the interpretation quite seriously (VICKERY & HOBBS, 2002). If it is unavoidable to measure in such an area, it is better for the profiles to be perpendicular to the pipes than
218
4.3 Direct Current Resistivity Methods
parallel to them. Another type of noise of natural origin, for example, induction effects of magnetic storms or atmospheric electrical discharges (sferics), can also significantly influence the measurements (HOOGERVORST, 1975). In many cases, the results of resistivity measurements may be considerably affected by near-surface conditions. It is important to take resistivity changes in the soil cover into account if it is necessary to compare a series of measurements (e.g., soundings) carried out at different times and meteorological situations. Variations in resistivity in the uppermost layer (caused by differences in near-surface moisture content as a result of alternating dry and wet periods, soil temperature, plant water uptake, etc.) can have a significant impact on the sounding curve and the results of the profiling. Another factor which adversely affects the results is topography (FOX et al., 1980). In a high-relief area, apparent resistivities cannot be calculated using the configuration factors given in Table 4.3-1 since these K values are for a horizontal ground surface. The terrain relief may have a strong influence on shaping the equipotential surfaces and, therefore, the K-factor has to be modified to take the terrain geometry into account. This is an important step in the interpretation of 2-D or 3-D resistivity measurements, which requires the topography along the profile to be surveyed by leveling. The results of leveling have to be taken into account as a first step in the 2-D modeling process. Mapping and profiling Mapping/profiling methods provide information about the lateral resistivity distribution within a certain depth range. For environmental problems they are mainly used to delineate the boundaries of concealed waste dumps or to locate contamination plumes. In the survey, an electrode array whose parameters are kept fixed is moved along profiles (profiling) or, if possible, on a regular grid (mapping). Thus, the investigation depth varies only slightly with changes in the subsurface resistivity distribution. Because of their operational advantages, Wenner, Schlumberger, and dipole-dipole arrays are the most commonly used (WARD, 1990; MILSOM, 1996). The geometric parameters of an array have to be adjusted to obtain the investigation depth necessary to achieve the project goal. These parameter values can be optimized using the results of several vertical electrical soundings carried out at well-chosen points in the survey area. The survey grid and the profile lines have to be established before starting the survey. One disadvantage of a mapping and profiling survey is the need for manpower – at least 3 to 4 people are necessary. Another problem is the long cables that have to be moved in the field to achieve large investigation depths. For this reason, electromagnetic methods have largely replaced resistivity profiling and mapping in recent years. These methods
Environmental Geology, 4 Geophysics
219
provide comparable results in many cases requiring less time and less personnel, i.e., they are cost-saving (see Chapter 4.4). An alternative to profiling is 2-D resistivity imaging. Although the costs is often almost the same as for conventional profiling, this method provides much more detailed information along the profile in both the lateral and vertical directions (BARKER, 1992; DAHLIN, 1996).
Vertical electrical sounding Vertical electrical soundings (VES) has been the most important tools in resistivity surveys for a long time, and for deep investigation depths (>100 m) they still are, although transient electromagnetic soundings (TEM) are being increasingly used for investigation of these depth ranges (see Chapter 4.4). VES is also used for near-surface environmental problems, but in this field 2-D resistivity measurements are becoming increasingly important. Soundings should preferably be used to investigate horizontally layered ground. They provide information about layer thicknesses and resistivity. The apparent resistivities allow conclusions to be drawn with regard to lithological parameters and the composition of the pore fluid (e.g., mineralization of groundwater). If the resistivity data can be correlated with drilling results or other a-priori information, the accuracy of the parameter values derived from 1-D inversion can be better than 5 %. Soundings should be carried out on an almost regular grid or along profiles. The result of one sounding on a profile can be used as input for the 1-D inversion of the next sounding on that profile. A geoelectrical crosssection can be produced from the results of the soundings. Because of its high vertical resolution, the Schlumberger array is preferably used for sounding. To obtain information about resistivity as a function of depth, the current electrode (C1, C2) spacing 1 L is increased stepwise (geometric sounding). It has proven to be useful to increase the current electrode spacing in logarithmic steps (optimum: 6 - 8 steps per decade), so that the points appear roughly equally spaced when plotted on loglog graph paper. The following sequence has proved useful: L/2 = 1, 1.3, 1.8, 2.4, 3.2, 4.2, 5.6, 7.5, 10, 13, 18, ... 100, 130, 180, ...1000 m. The potential electrode (inner electrodes) spacing l is increased (as long as the ratio l/L < 1/3) only if the measured voltage becomes too small and, therefore, the signal-to-noise ratio becomes too low. In this case, some of the measurements must be repeated (overlap readings), starting with the two current electrode spacings before the potential became too low. Overlapping segments of the sounding curve are obtained, which have to be shifted so that 1 For the electrode spacing L often the symbol AB is used.
220
4.3 Direct Current Resistivity Methods
in the overlapping range they are superimposed. This procedure is necessary because increasing the potential electrode spacing often causes the resistivity to change, due to inhomogeneity in the ground near to the potential electrodes or causes the coupling with the ground to change. Data quality should be checked already in the field by plotting the sounding curves on log-log graph paper. This way, the operator is able to react to data outliers, for example, resulting from insufficient galvanic coupling of the electrodes with the ground or geometric errors. The sounding sites (coordinates of the array centre, direction of the electrode spread) have to be documented in the field notebook. If there is rough terrain, the elevation of each sounding position has to be determined by leveling.
2-D resistivity surveys In the geoelectrical investigation of environmental problems, 2-D resistivity surveys (2-D imaging) have played an increasingly important role in the last few years. The advantages of 2-D measurements are their high vertical and lateral resolution along the profile, comparatively low cost due to computerdriven data acquisition, which means only a small field crew is needed (one operator and, for basic electrode spacings 5 m, one assistant). Owing to the performance capabilities of the available instruments, 2-D surveys are, in general, limited to investigation depths down to about 100 m. 2-D measurements are often the most suitable geophysical method for solving environmental problems, which are frequently related to 2-D or 3-D resistivity features. Many different instruments have been developed for 2-D measurements, usually multi-electrode systems, sometimes multi-channel ones. All systems use multi-core cables (see Section 4.3.4). The procedure for measurements with a Wenner electrode array is shown in Fig. 4.3-3. A multi-core cable with equidistant “takeouts” for connecting the electrodes to the cores is placed along the profile. The basic electrode spacing may be either the distance between the takeouts or a shorter distance. In any case, the electrodes should be positioned with equal spacings along the profile. For high quality data, it is important to ensure good galvanic coupling of the electrodes with the ground. Data acquisition is begun after the multi-core cables are connected to the switching box and the electrode contacts have been checked. The measurements are controlled by the microprocessor-driven resistivity meter or by a computer (BARKER, 1981).
Environmental Geology, 4 Geophysics
221
Fig. 4.3-3: Setup for a 2-D resistivity measurement with a Wenner electrode array
Several depth levels may be measured by increasing the electrode spacing stepwise as shown in Fig. 4.3-3. The apparent resistivities are plotted as a function of location along the profile and electrode separation. This 2-D plot is called a “pseudosection”. For the Wenner electrode array, the location on the profile is given by the center of the array and the depth (“pseudodepth”) is given by the spacing of the current electrodes. The “pseudodepth” corresponds to the depth of investigation (| a/2). Normally, two multi-core cables are used for each measurement. If the profile is longer than the total length of these two cables, the first cable is placed at the end of the second one and a new measurement is made. The survey can be carried out more efficiently if a third cable is available, as this could be laid out ahead while the measurement using cables 1 and 2 is being made. To starting and end points and the points where the profile changes direction define the profile. To avoid geometrical errors, these bends should not exceed 15 degrees. In rough terrain the elevation of several prominent points along the profile should be determined.
4.3.6 Processing and Interpretation of the Measured Data Mapping / profiling The mapping and profiling results are presented as colored contour maps of apparent resistivity or as profile plots, respectively. Profiling results should be shown as maps only if the distance between the profiles is not larger than five times the station interval along the profile, especially where resistivities vary considerably between the profiles. If there is rugged terrain in the survey area, a topographical correction is necessary before the final maps are made (FOX et al., 1980). The mapping data are interpreted qualitatively. If some resistivity
222
4.3 Direct Current Resistivity Methods
values in the area are known, the lithology in the survey area can be derived, lateral changes can be delineated (e.g., boundaries of concealed landfills), and information about contamination plumes obtained. Vertical electrical sounding The results of a resistivity sounding survey can be interpreted both qualitatively and quantitatively. Vertical electrical soundings can provide information about the areas between boreholes. If soundings are more or less regularly distributed in the survey area, apparent resistivity maps can be made for different electrode spacings C1C2. This type of map is similar to that obtained in a mapping survey, although the station spacings are commonly larger. In many cases they provide an idea about the resistivity distribution in the area. If soundings are carried out along profiles and the distance between soundings is not too large, the results can be displayed as contoured pseudosections of apparent resistivities. “Pseudodepths” correspond to approximately half the spacing of the current electrodes and, therefore, are preferably presented on a logarithmic scale. This gives a rough visual impression of the geological structures below the profile. The preferred quantitative interpretation of resistivity soundings requires information about the local geology and experience using the method, including knowledge of its limitations. Several computer programs are available for the inversion, allowing data processing, such as editing of data sets and sounding array parameters or shifting of overlapping segments to produce continuous sounding curves (see Section 4.3.5). Modern computer programs allow modelling of the individual segments in a Schlumberger sounding. The interpretation of a sounding curve to derive a layered resistivity model which fits the measured data well is carried out in three steps (ZOHDY, 1989; SANDBERG, 1993). In the first step, a reasonable starting model has to be created. Experienced geophysicists can derive the number of layers and an initial estimate of the resistivities and thicknesses of the layers directly from the sounding curve. The influence of layer resistivity and thickness on the curve shape is checked by doing both forward and inverse calculations. Important for the interpretation is that the program allows masking of noisy data points and incorporate of a-priori information about the local geology from well logs and other geophysical methods. The second step involves automatic inversion to confirm and refine the starting model. If model parameter values are known (layer thickness and/or resistivity), they can be held constant. If some parameters do not vary during the inversion, the remaining parameters are more reliably estimated. A leastsquares fit of the model curve to the edited data is then made until a selected fitting error is reached. Due to the equivalence basic (see above), several models may fit the sounding curve within the fitting error. Therefore, an
Environmental Geology, 4 Geophysics
223
inversion program should be able to determine the range of equivalence for a certain error threshold. Equivalence analysis makes it possible to generate a set of equivalent layered models that are also valid solutions of the inverse problem. They fit the data nearly as well as the best-fit model, but deviate from this within a defined error range. With this knowledge it is then possible in a third step to obtain a model which is geologically more plausible than the best-fit model, but which is still acceptable within the error limits. An example is given in Fig. 4.3-4. A by-product of the equivalence analysis is a „parameter resolution matrix“. This triangular matrix provides an indication as to how well the layer parameters (thickness and resistivity) are resolved. Owing to equivalence, the two parameters cannot be resolved independently for all layers (Section 4.3.3). The resolution matrix indicates when this may be expected (INTERPEX Ltd., RESIX-IP user´s manual, 1993). The best way to obtain a realistic interpretation of the soundings is to correlate soundings with lithological borehole logs or well logging data. As a rule, the interpretation is an iterative process because results of the individual soundings always have to be adjusted to the results of neighbouring soundings. The final result of a quantitative interpretation can be presented as a geological model (cross-sections) derived from the sounding data. Contour maps of the depth to layer boundaries or of the thickness of different layers can be constructed from the results of several soundings regularly distributed in the survey area. It has to be mentioned that such layer boundaries do not necessarily correspond to lithological boundaries - they represent significant changes in the electrical properties. This can also occur within a lithological layer (e.g., a change in the mineralization of the groundwater in an aquifer).
Fig. 4.3-4: Graphical representation of equivalent models. Each of these models fits the measured data equally well.
224
4.3 Direct Current Resistivity Methods
2-D resistivity surveys For many environmental problems it is preferable to carry out 2-D resistivity measurements instead of simple soundings and/or profiling. The aim of a geoelectrical survey at the Earth surface is to provide detailed information about the lateral and vertical resistivity distribution in the ground. As the principles applied are similar to those of computer tomography in medicine, such geoelectrical investigations are also called impedance tomography or 2-D resistivity imaging. The first – qualitative - result of a 2-D resistivity survey is a pseudosection along the profile (EDWARDS, 1977). Pseudosections display the apparent resistivity as a function of location and electrode spacing, which is indirectly, related to the depth of investigation of the array (ROY & APPARAO, 1971; BARKER, 1989; SHERIFF, 1991). Figure 4.3-3 shows the principle behind the generation of a Wenner pseudosection. Pseudosections provide an initial picture of the subsurface geology. A 2-D inversion of the measured data is necessary for the final interpretation. This process transforms the apparent resistivities and “pseudodepths” into a 2-D model. In recent years, several computer programs have been developed to carry out such inversions (SHIMA, 1990; BARKER, 1992; LOKE & BARKER, 1995 and 1996). Depending on the algorithm used, the result is a smoothed layer model or a block model showing sharp layer boundaries, comparable to the result of 1-D inversion. The model used for inversion consists of a number of rectangular cells (blocks). The size of these cells is determined either automatically as a function of the electrode spacing, or manually by the user. In general, the size of the cells increases with increasing depth and towards the beginning and the end of a profile, due to decreasing spatial sensitivities in these areas (see Fig. 4.3-2). All inversion programs require a reasonable starting model (initial estimate). This model is either determined by the program itself or user defined (BARKER, 1989; ZOHDY, 1989). The final model is calculated in an iteration process which includes step-by-step forward modeling and inversion. The program compares the forward modeling results with the measured data, the quality of the fit is determined and the parameter values for the next iteration step are specified. This process is continued until the maximum number of iterations or a selected deviation (error) is reached. The modeling program must allow interactive modification of parameters. For rough terrain, the program should be able to include topographic correction of measured data. Recent programs allow the user to place the electrodes in non-standard positions (e.g., non-regular spacing, electrodes in boreholes or underwater). As in the 1-D inversion of vertical soundings it is advantageous if a-priori information can be included.
Environmental Geology, 4 Geophysics
225
In spite of the various capabilities of different programs, the final assessment of inversion results has to be done by the user, i.e., the geophysicist. Due to the limited number of values and precision of the data, all results are ambiguous (Section 4.3.3). This ambiguity can only be reduced by including all available information from boreholes and other geophysical surveys. As the inversion will not normally provide sharp layer boundaries but resistivity gradients (smooth inversion), data for a vertical electrical sounding in areas with almost horizontal layers can be extracted from the 2-D data set and inverted as 1-D vertical soundings. Generally, if gradual structural transitions are indicated by gradual transitions in the resistivity cross-section, a smooth 2-D inversion should be carried out; if the structures are known to have sharp boundaries, then block inversion should be employed. The final result of a 2-D resistivity survey is a cross-section of the calculated rock resistivities along the profile line. This cross-section should include the structural interpretation of the resistivity data (e.g., the groundwater table or the boundaries of a waste dump). If the spacing between profiles is not too large, horizontal resistivity sections for different depths can also be derived from the data.
4.3.7 Quality Assurance The following measures are important for quality assurance in a dc resistivity survey. During the survey the following aspects should be checked in the field: x coupling of the electrodes to the ground (contact resistance), x drifting self-potentials, x the parameters (voltage, current) employed for the measurement, x stability and precision of the measurements (e.g., by repeated readings), x leakage currents from large power consumers, e.g., industrial plants or electric railways (industrial noise), x influence of electromagnetic coupling if long cables are used, x influence to topography, and x the influence of meteorological factors.
226
4.3 Direct Current Resistivity Methods
The field notebook documentation should include: x the instrument employed and the parameters used for the measurements (array type and geometry), x a topographic map (including major topographic features, such as roads, buildings, rivers and lakes) showing the location of profiles and measurement stations; this description shall enable the reestablishment of the survey grid, x information about the location of buried objects (cables, metal pipes, etc.), x rapid changes in the relief, topographic survey results, x information about visible sources of anomalies (e.g., iron fences, waste dumps, wet and dry areas, changes in the composition of the ground surface), and x weather conditions. Attention must be paid to the following aspects of data processing and interpretation: x The raw and processed data must be delivered to the client in digital form. It must also be archived so that it is available for later reprocessing. x The client must be provided with a description of the processing parameters and software (e.g., for data correction, data filtering and/or smoothing). x A technical report should be submitted as part of the final report. x The results have to be presented in a form that can be understood by the client. The presentation must also provide an explanation of uncertainties in the interpretation and recognized problems in the modeling, particularly from a geophysicist’s point of view.
227
Environmental Geology, 4 Geophysics
4.3.8 Personnel, Equipment, Time needed Personnel mobilization and demobilization
Equipment
Time needed
depending on the distance to the survey area 1 4WD vehicle or van
mapping: Schlumberger P1P2 / 2 < 3 m
1 operator, 2 assistants;
resistivity meter, cables, electrodes
C1C2 / 2 < 30 m: 500-800 N/d
P1P2 / 2 > 3 m
1 operator, 3 assistants;
C1C2 / 2 > 30 m: 300-500 N/d
Wenner C1C2 > 15 m soundings (VES): Schlumberger C1C2 < 130 m
1 operator, 3 assistants
400-700 N/d
resistivity meter with appropriate power supply 1 operator, (high capacity for deep 1-2 assistants; VES), cable reel, 1 operator, electrodes, (optional C1C2 t 130 m 2-3 assistants walkie-talkie) 2-D resistivity survey: resistivity meter with switching unit, multi-core (imaging, tomography) basic electrode spacing: cables, passive or active electrodes, 1 laptop, a Ic = arcsin (1/ H r* ), even for real values of the permittivity. This means that there is a phase change of the transmitted and received wavelet for these transmission angles. If the transmitting and receiving antennas are identical, the square of the single antenna pattern TA (I) = t A2 (I) gives the amplitude and phase patterns of the system as a whole.
Wave paths, traveltimes, and amplitudes The propagation of radar waves can be described by an idealized representation of rays as in optics and seismics. A simple two-layer model for the GPR method requires four wave paths and traveltime curves. The GPR ray path scheme is shown in Fig. 4.5-5. Two direct waves with different phase velocities and amplitudes travel along the ground surface: the air wave and the ground wave (BAÑOS, 1966; CLOUGH, 1976). Since the air wave travels with the largest possible velocity for electromagnetic waves – the velocity of light in a vacuum – it can be used to determine time zero (like the time break in seismics). The velocity in the uppermost stratum is determined from the ground wave. Changes in the direct waves indicate changes in the uppermost stratum (e.g., moisture content, type of rock).
Environmental Geology, 4 Geophysics
293
Fig. 4.5-5: GPR ray paths – principle sketch
The traveltimes of the air wave ta and the ground wave tg are given by ta
x c0
and
tg
x , vg
where x is the distance from transmitter to receiver dipole [m], va is the velocity of the air wave (§ c0 § 0.3 m ns-1), and vg is the velocity of the ground wave [m ns-1]. Like in reflection seismics, the traveltime tr of reflected waves is given by the hyperbola: tr
1 v
x 2 4h 2 .
(4.5.9)
Because v is always less than c0, a lateral wave is generated at the critical angle Ic = arcsin (v/c0), analog to the head wave in refraction seismic. This wave propagates in air parallel to the ground surface. The critical angle Ic is related to the critical distance xc, which is given by xc
2hv c 02
v2
.
294
4.5 Ground Penetrating Radar
Fig. 4.5-6: Traveltime diagram for the horizontal two-layer case
The traveltime of the lateral wave is given by tl
1 1 x 2h 2 2 . c0 c0 v
Refracted waves are seldom observed, because in most cases the velocity decreases with depth. A traveltime diagram of the different wave types is shown in Fig. 4.5-6. The field strength of the direct wave Ea (tangential component parallel to the dipole) decreases in the far-field with the square of the distance from the source. The field strength of the ground wave Eg is diminished in addition by absorption and scattering. Ea |
1 x2
Eg |
and
1 exp D x . x2
The field strength of the lateral wave (analog to seismics) is given by (BREKOVSKIKH, 1980) El | x
1 x xc
3
for
x ! xc ,
and that of the reflected wave in the far-field by
295
Environmental Geology, 4 Geophysics
E r | exp D s s 1 r I TA I
where
s
I A
D
G r TA(I)
AGrT ,
(4.5.10)
x 2 4h 2 ), is the path length ( s is the angle of incidence (I = arctan (x/2 h)), is the absorption und scatter (frequency dependent), is the attenuation coefficient [m-1], is the geometry, spherical divergence, is the reflection coefficient for plane waves derived from the amplitudes, and is the antenna pattern.
The factors of proportionality A, G, and T are derived from the field strengths in the near-field of the transmitter dipole and the modification of the antenna pattern by coupling and shielding. In the case of diffractors, the following equation (e.g., JANSCHEK et al., 1985) is a suitable alternative to Equation (4.5.10) in the zero-offset case (x = 0) for waves diffracted back towards the source: G 2O 2 (4.5.11) exp 4D h , Pr Pt q s 3
4S
where
Pt Pr qs G
O h
h4
is the transmission power [W], is the power of the field at the receiver [W] is the scatter cross-section (area of the first Fresnel zone) [m2], is the antenna gain (relative to that of a spherical dipole), is the wavelength of the center frequency [m], and is the distance between the antenna and the diffractor [m].
Vertical and horizontal resolution Resolution is a measure of the ability to distinguish between signals from closely spaced targets. In ground penetrating radar the resolution depends on the center frequency (or wavelength, which is proportional to the pulse period) and the bandwidth as well as on the polarization of the electromagnetic wave, the contrast of electrical parameters (mainly conductivity and relative permittivity), and the geometry of the target (size, shape, and orientation). Important are also the coupling to the ground, the radiation patterns of the antennas (especially the diameter of the first Fresnel zone 7 ), and the noise conditions in the field. As a rule of thumb, the vertical resolution is theoretically one-quarter of the wavelength O = v f--1, where v is the velocity of the electromagnetic wave in the medium (see Table 4.5-1) and f is the center frequency. 7 See Section 4.6.3.5 Equation (4.6.8) and Fig. 4.6-7.
296
4.5 Ground Penetrating Radar
Example: The wavelength O is equal to 1.2 m for a limestone with v = 0.12 m ns-1 and a 100 MHz antenna (1/f = 10 ns); the resolution will be on the order of 0.3 m. For very good conditions, the resolution can be one-tenth of the wavelength and for unfavorable conditions one-third of the wavelength. Because the velocity is lower in wet rock than in dry rock (e.g., sand, see Table 4.5-1), the resolution will be better in wet materials. Due to the abovedescribed effects in real Earth materials, the actual resolution will be in the order of one wavelength. The horizontal resolution is thoroughly discussed in Section 4.6.3.5 for the case of reflection seismics. It has been shown that objects can be separated laterally if their distance is larger than the diameter dF of the first Fresnel zone. For example, a GPR system operating at fm = 200 MHz in an environment with v = 0.1 m ns-1 has a wavelength of O = 0.5 m. The horizontal resolution at a depth h = 4 m will be dF = (2 O h)1/2 = 2 m, which is much poorer than the vertical resolution of better than 0.5 m.
Estimation of penetration depth and amplitude To avoid dispersion, the operating frequency should be chosen so that tan G < 0.5 (see Equation 4.5.3). Solving Equation (4.5.3) for frequency, we obtain fm t
where
36 000
U H r'
,
is the center frequency [MHz], is the electrical (dc) resistivity [:m], and is the real part of the relative permittivity (see Table 4.5-1). If no other information is available, the value of the absorption coefficient D' can be taken from Table 4.5-1 or, using Equation (4.5.3), estimated from the resistivity U provided by dc resistivity measurements or taken from Table 4.3-2 (Chapter 4.3). The values for the parameters Z and H cr are inserted in Equation (4.5.4), with 0.5 for tan G. Equation (4.5.4) then reduces to
Dc | where
fm
U H cr
1640
U H cr
D'
,
(4.5.12) is in dB/m.
For successful application of the method, wave absorption along the way from the ground surface to the reflector and back to the surface should not exceed 40 - 60 dB. Experience has shown that otherwise spherical divergence, reflection and dispersion loss, geological and technical noise will take up the remaining dynamic range. For example: a center frequency > 40 MHz is
Environmental Geology, 4 Geophysics
297
selected for ground with a resistivity of 90 ȍm and H cr = 10. A value of 5.8 dB/m is obtained for the absorption coefficient using Equation (4.5.12). Thus, the maximum depth of penetration is about 5 m (hmax § 60 dB/2Į'). A more exact value can be obtained using Equations (4.5.10) and (4.5.11) if the system parameters (bandwidth, sensitivity, output of the pulse generator, antenna efficiency, effective area of the antennas, spectral distribution of the noise, and the high-frequency properties of the ground) are known.
4.5.4 Instruments GPR systems consist of a pulse generator, an antenna for transmission of highfrequency electromagnetic waves, a second antenna to receive the direct and reflected impulses or a switch for switching between transmission and reception if only one antenna 8 is to be used, and a receiver which converts the received signals to be recorded and displayed. These components are designed and arrayed differently by different producers of radar equipment, but the functionality is generally the same.
Antennas Broadband antennas are needed to transmit and receive short electromagnetic impulses. Conventional broadband systems with directional antenna gain, as used for radio and television reception in the VHF and UHF ranges (e.g., logarithmic-periodic antennas), are unsuitable for single pulses. A broad bandwidth is generally achieved by damping electrical dipoles. Several antenna designs have proved to be useful, e.g., linear dipoles with hyperbolic resistive loading (see WU KING, 1965 for theoretical treatment) and combinations of butterfly and loop dipoles with empirically determined resistivity damping (Fig. 4.5-7). Antennas with loading according to WU KING have a signal loss of 20 dB relative to an undamped dipole. The radiated wavelet is approximated by the time-derivative of the excitation function. The frequency spectrum of a damped loop dipole is narrower, the waveform is similar to a Ricker wavelet (Section 4.6.3.3), and the amplitude loss per antenna pair is about 10 dB.
8 The use of a single antenna for both transmission and reception is called monostatic radar;
the use of two antennas is called bistatic radar, with the results assigned to the midpoint between the two antennas. Most GPR investigations can be carried out with monostatic or bistatic systems. For some applications, e.g., wide-angle reflection and refraction (WARR) a bistatic system is necessary.
298
4.5 Ground Penetrating Radar
(a)
(b)
Fig. 4.5-7: Schematic of examples of broadband antennas: (a) butterfly dipole, ROTHAMMEL (1991); (b) V-dipole with resistive loading, e.g., BLINDOW (1986)
When radar data is interpreted it is necessary to keep in mind that the source pulse is longer than one wavelength and may have a complex waveform. The downgoing wavelet is modified by ground coupling and attenuation effects9 . Therefore, the reflection is also complex. It consists of more than one wavelet. Typical antenna patterns are shown in Fig. 4.5-4 for different dipole orientations. When measurements are made within buildings, under power lines and trees, etc., “air reflections” are obtained from reflectors and diffractors in the half-space above the antennas. To suppress these disturbances, absorber and/or metal screens can be placed above the antennas, especially at center frequencies above about 100 MHz (i.e., when compact dipoles are used). When these are used, however, the antenna patterns are changed. The antennas are usually completely enclosed and the manufacturer usually gives only the central frequency (which is usually for use in the air and thus too high for georadar) and a note that a screening effect will take place. Borehole antennas often are due to the limited space simple or damped dipoles. To obtain a directional effect, the dipole is exentrically embedded in a dielectric material; another possibility is to use a loaded loop antenna.
Pulse generators Pulse generators for producing short, high-energy impulses for use in the field are often build up as cascade generator with transistor switches using the avalanche effect 10 . In principle, a series of parallel capacitors are charged and then discharged to the antenna in a cascade by the transistors switches (PFEIFFER, 1976). Pulses with nanosecond rise times and amplitudes of up to 2 kV can be produced with frequencies of more than 100 kHz. The typical pulse length of a transmitted electromagnetic wave is < 20 ns, depending on 9 Attenuation is the process of transformation and loss of energy of the propagating electromagnetic wave. The energy loss through geometric effects in the wave propagation, such as scattering, geometric spreading, multipathing, etc., is called apparent attenuation. 10 Avalanche effect is a process in solids analog to gaseous discharge where few injected electrons release many further electrons like an avalanche.
Environmental Geology, 4 Geophysics
299
the antenna frequency and type. It is important that the transmitted signal is generated and repeated with a high accuracy. The timing accuracy within radar systems used for geophysics is usually ±1 ns. The repetition rate is typically 50 000 times per second. For measurements with small offsets it is important that no further signal is produced by the pulse generator between pulses. Such signals would appear as arrival times with either constant or variable traveltimes. Recently developed MOS switches and other high-voltage devices now provide alternatives for pulse generation.
Receiver systems The time window for georadar measurements ranges from several 10 ns (for travel paths of several meters) to 50 µs (e.g., for travel paths of 4 km in ice). Depending on the frequency used, the measurements are made with sampling intervals of down to less than nanoseconds. Either analog or digital systems with a high dynamic range can be used. For reasons of weight and power consumption, mobile systems are usually designed to take only one sample for each transmitted pulse as the antennas are drawn along the profile (i.e., sequential measurements). The principle is discussed, for example, by PFEIFFER (1976). This is not a disadvantage for continuous measurements, since achievable pulse rates are very high and the pulse generators normally have a long lifetime. The sequentially received high-frequency signals are converted to audio waves so that they can be digitized and recorded. The achievable dynamic range of good sampling systems is 80 - 90 dB. The sensitivity (without stacking) under ideal conditions depends on the thermal noise in the input and the noise factor of the receiver. The effective thermal noise potential Ueff (also called Nyquist noise) is given by the following equation: U eff
4kT R' f
where
k T R
'f
is the Boltzmann constant (1.380510-23 A s K-1), is the absolute temperature [K], is the antenna impedance (electrical resistivity) [:], and is the band width [Hz].
Example: For R = 50 :, T = 300 K, and ǻf = 400 MHz, the thermal noise potential Ueff = 36 µV. Thus, the resolution of the digitizing system should be about 10 µV. The maximum range is then about 0.65 V for a 16 bit system, which is sufficient in most cases. To avoid oscillation of the system as a whole, the antennas must be matched to the input of the receiver, and propagation of high-frequency electromagnetic waves in metal cables must be prevented, especially between
300
4.5 Ground Penetrating Radar
the transmission and receiver antennas. The GPR systems of some manufacturers use fiber optic cables for this reason. High-quality digital storage oscilloscopes (DSOs) and other fast A/D converters make it possible to replace the manufacturers “black box” analog sampling systems with systems with known specifications. Because these digital systems require only one “shot” per time trace, it would be possible to use high output pulse generators with a lower repetition rate. For surveys of glaciers, there are a number of prototype digital systems available (e.g., JONES et al., 1989; WRIGHT et al., 1990). Since most systems have only one to four channels, multiple coverage measurements like those in reflection seismics are done sequentially. Normally, georadar measurements are carried out with single coverage at contant antenna offset.
Performance factor The system dynamics is given by some manufacturers as the ratio of the peak voltage of the pulse generator and the minimum recordable voltage of the receiver in dB (performance factor PF). Performance factors of 130 to 140 dB are typical for existing systems. Because antenna efficiency, ground coupling, and spectral content of the broadband transmitter pulse are not specified, the PF is of little use in practice. It could not in any case be used to calculate the depth of penetration, etc. More important for good results are the dynamic range of the receiver, adaptation of the pulse generator and antennas to the ground material, clean excitation pulses, and avoidance of high-frequency oscillations in the cables.
4.5.5 Survey Practice There are several procedures used for GPR surveys: radar reflection profiling, wide-angle reflection and refraction (WARR), common-midpoint soundings (CMP), and radar tomography. See Section 4.5.10 for special applications. For radar reflection profiling one or two antennas are moved over the ground while simultaneously measuring the traveltimes of the reflected radar pulses, as is done in seismic reflection profiling (see Chapter 4.6). This method is the most frequently used for GPR surveys. For WARR soundings, the transmitter is kept at a fixed location and the receiver is moved away from it. For CMP soundings, both the transmitter and receiver are moved simultaneously away from a fixed midpoint. This makes it possible to determine the velocity-depth function. WARR/CMP soundings can be carried out only with bistatic antenna systems. For radar tomography (trans-illumination) applications, transmitter and receiver antennas on opposite sides of the volume to be investigated (e.g.,
Environmental Geology, 4 Geophysics
301
between boreholes or for the non-destructive testing of walls and pillars) are moved as shown in Fig. 4.5-13. As for all geophysical methods, the objective of the field work must be clearly defined for the planning and execution. Important for an assessment of success of the method are the depth to the groundwater table, the kind of cover sediments (good-conducting, cohesive material or low conductivity, noncohesive material), and the location of buildings, buried and above-ground cables and other buried objects that are not investigation targets. Visible objects should be marked on the base map, as well as distances and dimensions. When ground penetrating radar measurements are interpreted, the influence of temperature, precipitation and chemicals present must be taken into account. Therefore, it is necessary to note the weather conditions and any observations of pollution during the measurements. The suitability of an area or target object for georadar can be checked with a test measurement or by geoelectric sounding, from which the absorption can be estimated from the electrical resistivity of the ground. Unsuitable are areas with moist clay and those paved with slag. To calibrate the GPR data, test measurements should be made along a profile with known underground conditions that are typical of the area. The target object (buried object, reflection horizon) should be recognizable in the test data. The operating frequency (and thus the resolution), the distance between the transmission and receiver antennas (constant offset for simple coverage measurements), and the distance between profiles and measurement points must be selected on the basis of the objectives. The useable dynamic range is negatively influenced by a small offset; if the offset is too large for investigating shallow depths, direct and lateral signals will overlie the reflections. As a rule, the offset should be about a fourth to a fifth of the expected depth to the reflector or diffractor. The spacing of the measurements should be a compromise between the necessary resolution and the fastest possible speed of measurement. For measurements of an entire area, the guidelines for reflection seismic should be used (see Chapter 4.6). All GPR systems provide possibilities for filtering the data during acquisition in order to sharpen the signal waveform. Both high-pass and lowpass filters are used for this purpose. A rule of thumb given by REYNOLDS (1997) says that the filter settings should kept as broadband as possible so that potentially valuable signals are not removed during the acquisition phase. Digital systems have gain-setting options and a stacking function to optimize data quality.
302
4.5 Ground Penetrating Radar
4.5.6 Processing, Presentation and Interpretation of the Measured Data Filtering is the first step in post-acquisition data processing. For many applications, this is sufficient to prepare the data for presentation and interpretation. Data-processing packages are available for a more thorough analysis. Most of the steps for a seismic survey can be directly applied to a georadar survey. Normally, however, multiple coverage measurements are seldom made. Hence, CMP stacking, one of the most important processing steps in seismics, is not usually done. There are also differences in the application of deconvolution. The complex reflection and transmission coefficients and dispersion effects during propagation cause wavelet deformation. Because the wavelets are not minimum-phase, but have a "mixed-phase" character, the deconvolution method used in seismics cannot be applied directly. A data adaptive deconvolution is necessary in GPR. Radargrams often show a complicated picture, due to a large number of diffraction hyperbolas, which require considerable experience to interpret. Usually the location of the diffraction center is indicated by the apex of the hyperbola (e.g., the location of a buried pipeline or cable, Fig. 4.5-8). If there is little absorption, the migration programs used for reflection seismic can be used to obtain a good mapping of diffractors with considerable improvement in resolution. If there are reflection horizons in the survey area, the velocity-vertical traveltime function can be obtained from CMP or well logging measurements, analogously to reflection seismic. The analysis of diffraction hyperbolas is suitable only for a rough estimate of velocity, since location and shape of the diffractor are often not sufficiently known. An exception is diffraction hyperbolas from profiles perpendicular to buried pipelines or cables. These velocities can be used to calculate depth from traveltimes. Due to the high data density from profile measurements, the data are often displayed compressed in grey-scale or raster plots instead as individual data points. In the presentation of GPR data it is necessary to give the traveltime and a distance scale, as well as the time-zero point, the antenna offset, the velocity-depth function, and/or a depth scale derived from it. The identification of the horizons and objects should be well founded. To avoid erroneous interpretation by the client, system noise (e.g., cable waves, which appear in the radargram as parallel stripes at constant time intervals) should be labeled. There are two types of areal surveys: x Use of profile spacing larger than half the wavelength of the central frequency for reconnaissance surveys: Such a survey is suitable for following subhorizontal horizons or to map areas with varying cover layer properties. The data is processed like for single profiles and presented on a
Environmental Geology, 4 Geophysics
303
map, e.g., horizon depth relative to MSL or special features in the cover layer. x 3-D measurements (without spatial aliasing 11 ): The amount of data, the necessary processing steps, and the possibilities for representation are comparable to 3-D seismics. Special programs are available for processing the data. In some cases it is difficult to distinguish between significant reflections and multiple events, extraneous reverberations, off-section ghosts, etc. In such cases the interpretation of georadar data can be checked by modeling. Appropriate modeling software has been developed for the 2-D and 3-D cases (CAI MCMECHAN, 1995; GOODMAN, 1994). The structural modeling in these programs is done by ray tracing according to optical principles. The dynamic aspects of wave propagation (energy conditions and signal form) influenced by absorption, divergence, reflection, and transmission are usually taken into consideration in these programs. Calculations using the FDTD (finite-difference time-domain) method have become more powerful during the last decade (e.g., RADZEVICIUS et al., 2003; LAMPE et al., 2003).
4.5.7 Quality Assurance General guidelines for quality assurance are given in Chapter 2.6. For georadar, the following points should be observed: x Diffractors and reflectors (trees, powerlines, roofs, etc.) above the antennas must be documented. x Recognizable changes in the cover layer and indications (e.g., manhole covers and hydrants) of buried pipelines or cables should be documented. Cable detectors should be used. x Survey parameter values appropriate to the survey objectives (operating frequency, antenna offset, location of the profiles, and orientation of the antennas relative to the profiles, especially when pipelines or cables are the target) should be selected. x All survey parameter values and equipment settings should be recorded in the field and deviations from normal documented. Profile position and parameter values not recorded in file headers must be noted during the survey in files for this purpose.
11 Spatial aliasing occurs when the data density is too small and can lead to overlook of
local anomalies.
304
4.5 Ground Penetrating Radar
x If the measurements are affected by disturbances in the functioning of the apparatus, or erroneous or nonoptimal parameter values are obtained, the profile is to be resurveyed before continuing the survey. x The kind of calculations used to estimate velocities for the depth calculations are to be noted: CMP (comparable to WARR = wide-angle reflection and refraction), diffraction hyperbolas, well logging, measurement of the permittivity of samples, calibration with respect to known reflectors or objects (e.g., from exposures), or estimation from experience. x The data is to be delivered to the client in well documented form as legal evidence and for possible reinterpretation together with the results of other methods.
4.5.8 Personnel, Equipment, Time Needed Personnel mobilization and demobilization topographic survey field work
Equipment
Time needed
depends on the distance to the survey area see Chapter 2.5 1 geophysicist (or technician) 1 assistant
1 4WD vehicle or station wagon, 1 georadar system, various types of antennas (in pairs) and accessory equipment (tape measure, survey wheel, or other instrument for measuring distance, etc.) data processing, 1 geophysicist PC or workstation with interpretation, > 256 MB RAM and (1 or 2 and report > 40 GB hard disk; radar assistants) preparation or seismic software, printer (plotter) a is the distance between measurement points, d = 10-h workday
for a = 0.02 - 1 m, 200 - 20 000 m d-1, depending on the conditions in the field, the objectives, and the type of antennas
1 - 2 days are necessary for each day in the field
Environmental Geology, 4 Geophysics
305
4.5.9 Examples Localization of objects Georadar can be used to localize both metallic and nonmetallic pipelines, cables, and other objects. Possibilities for application exist also in the field of archeology and in non-destructive testing. The size of the target is normally smaller than the wavelength of the radar wave, so that characteristic diffraction curves are obtained. The shape of the diffraction curves from pipelines and cables depends on the direction of polarization of the propagating wavefield and the properties of material it passes through. Due to changes in propagation velocity, backfill, e.g., in cable and pipeline trenches, creates additional pull-ups or push-downs 12 , which can indicate the presence of the object. The resolution of closely spaced pipes is not very high. Cohesive or inhomogeneous ground negatively affects the determination of the location of an object. The antennas must be located near (about 10 cm) or at the ground surface. The interpretation of a radargram is based mainly on the regocnition of diffractions. 3-D measurements have proved useful for complicated cases, e.g., where linear targets cross each other. An example is shown in Fig. 4.5-8, in which the locations of the pipelines can be clearly seen, marked by arrows. The section A - A' is running perpendicular to pipe L, which is indicated by a diffraction hyperbola. Section B - B' extends along pipe L, indicated by reflections. The kind of pipe (metallic, plastics) can be identified by analyzing the intensity of reflections and from previous knowledge. A second example is given in Fig. 4.5-9, which shows buried vault structures below a church.
12 Pull-ups or push-downs are sudden rises or depressions in a reflection event. They are
observed when there is a large difference between the velocities inside a buried object and the surrounding material (or the air/soil boundary in the case of a cavity) and may not be interpreted as faults.
306
4.5 Ground Penetrating Radar
Fig. 4.5-8: Radargrams showing the location of pipes, left: section A - A', right: section B - B' perpendicular to A – A', GGU, Karlsruhe, Germany
Fig. 4.5-9: Radargrams showing the location of buried objects, A: top edge of a vault, B: bottom edge of the masonry, C: object in the cavity, D: bottom of the cavity, GGU, Karlsruhe, Germany
Environmental Geology, 4 Geophysics
307
Quaternary geology and hydrogeology The location of the groundwater table can often be clearly observed in radargrams from coarse sands, gravel and porous limestone layers, caused by the large difference in impedance between unsaturated and saturated rock. Reflections are observed at layer boundaries when both layers have distinctly different water contents. The depth of penetration in saturated rock depends strongly on the conductivity of the water – from a few decimeters to several tens of meters. In the example in Fig. 4.5-10, the base of the aquifer at 28 m depth, mapped in parts of the profile, was confirmed by drilling. Survey setup: system antennas sampling interval measuring speed personnel processing
sampling system, opto-electronic transmission and 2 kV pulser (developed by the University of Münster), dipoles with resistive loading, approx. 50 MHz, 1 ns, 20fold stacking, 3 km h-1, wheel triggering, 5 traces m-1, 1 operator, 1 assistant, frequency filtering, time-dependent amplitude control.
Fig. 4.5-10: Radargram showing the groundwater table, layer boundaries, and the base of the aquifer in glacial sediments in the Lüneburg Heath, Institut für Geophysik, Westfälische Wilhelms-Universität Münster (1992)
308
4.5 Ground Penetrating Radar
Delineation of a mineralized fault in consolidated rocks GPR has a large penetration depth in consolidated, unweathered, dry rocks, like dolomite due to high electrical resistivity. In contrast, mineralized, clayfilled, and water-bearing faults have electrical properties (resistivity and permittivity) that are different from those of the surrounding rocks. Such faults are good reflectors for the radar waves. Fig. 4.5-11 shows a radargram recorded during the investigation of a fault containing a lead and zincblende mineralization in a dolomite mine. The depths were calculated from the traveltimes of the reflected signals and the velocity of electromagnetic waves in the dolomite. The velocity was determined in two ways: from a CMP sounding and from diffraction hyperbolae.
Fig. 4.5-11: Radargram recorded in an investigation of a mineralized fault in a dolomite, frequency 100 MHz, GGU, Karlsruhe, Germany
Environmental Geology, 4 Geophysics
309
Investigation of a domestic waste site in a refilled open-pit mine Georadar measurements were performed along with geoelectric resistivity measurements in the area of exhausted open-pit lignite mines. An example of these measurements on the surface of a refilled pit is shown in Fig. 4.5-12. When pumping to keep the open pit dry was switched off, the groundwater table rose in the stripped overburden filling the pit. Beginning at the edge of the pit at the left, the reflection from the water table can be clearly seen. Between 160 and 215 m from the edge of the pit, the location of a covered domestic waste landfill in a hole in the refilled overburden can be recognized in both the radargram and the resistivity curve. The base of the waste – dumped without a base seal – is below the water table, making it possible for pollutants to enter the groundwater and eventually surface water bodies. The measurements were made in continuous mode from a 4WD vehicle. Survey set up: system frequency record length scanning rate speed personnel
SIR-8 (GSSI), 80 MHz, 160 ns, 25.6 s-1, 5 km h-1, 1 operator, 1 assistant.
Fig. 4.5-12: Domestic waste landfill at the edge of an exhausted lignite open-pit mine, Lausitzer Braunkohlen AG, Arbeitsgruppe Geophysik
310
4.5 Ground Penetrating Radar
Radar tomography to assess the ground below buildings Radar velocity tomography in boreholes can be used to assess foundations, even below existing buildings. In gypsum and other evaporates, solution processes (subrosion) change the properties and structure of the ground. The objective of the site survey was to investigate the rock structures and properties below an old church. Fig. 4.5-13 shows the measurement scheme used in radar velocity tomography. Transmitter and receiver are placed in separate boreholes and the traveltime of the radar wave is determined for each raypath. The velocity distribution (Fig. 4.5-14, left) can be calculated from the traveltime and the known distance using a tomographic inversion algorithm. The geological cross-section (Fig. 4.5-14, right) was derived from the velocity distribution of the radar waves. The cross-section contains information about the structure and the rocks in the ground below the church.
Fig. 4.5-13: Radar velocity tomography, ray pattern for several transmitter and receiver positions between two boreholes, HALLEUX & RICHTER (1994)
Fig. 4.5-14: Radar velocity tomography, left: velocity of the radar waves in the ground between two boreholes, BK1 and BK2, determined using the measurement scheme shown in Fig. 4.5-13. Right: geological cross-section derived from the radar data, HALLEUX & RICHTER (1994)
Environmental Geology, 4 Geophysics
311
312
4.5 Ground Penetrating Radar
Investigations of concrete constructions Ground penetrating radar is an excellent method for quality control of concrete constructions (Fig. 4.5-15). Defects in concrete usually cannot be concealed by rebars if the profiles cross the rebars (steel rods).
Fig. 4.5-15: Radargram recorded for quality control of concrete constructions, cube-shaped voids (F1) and an air gap (F2), gravel nests at various depths (F3 and F4), uncorroded reinforcement bars (rebars) (B) and corroded rebars in a gravel nest (F5) in a concrete plate, GGU, Karlsruhe, Germany
Environmental Geology, 4 Geophysics
313
Examination of masonry structures Retaining walls are often used to stabilize embankments. But moving water or other processes (e.g., burrowing animals) change the structure and destabilize the construction. Fig. 4.5-16 shows an example of a GPR investigation of a retaining wall. Several parallel, vertical profiles were measured on the wall. Due to the wetness and salt content of the masonry, a 200-MHz frequency antenna was used. The thickness of the wall, structures in the wall, as well as cavities were determined by the measurements.
Fig. 4.5-16: Investigation of a retaining wall, left: photo from the measurements, middle: radargram of a vertical profile, right: section showing the result of the interpretation of the GPR data calculated using v = 0.13 m ns-1; H cavity, A inhomogeneity, R interior side of the wall, S internal boundary within the wall, GGU, Karlsruhe, Germany
Investigation of residual foundations Residual foundations impedes the reuse of sites. GPR measurements on a grid of profiles or several parallel profiles with small line spacing are useful to reveal construction obstacles in the ground. For an areal representation, the strength of radar signal amplitudes is estimated on all profiles from the radargrams for a given traveltime. From this, a time slice representation (Fig. 4.5-17) is plotted. The time slice can be assigned to an approximate depth. Dark shading indicates high signal amplitudes from residual foundations.
314
4.5 Ground Penetrating Radar
Fig. 4.5-17: Areal representation (time slice, plan view) of radar signals backscattering for a depth of about 1 m below ground surface, GGU, Karlsruhe, Germany
Identification of organic contaminants To determine the effectiveness of GPR for identifying organic contaminants, a test was carried out in 1991 by the Canadian Waterloo Center for Groundwater Research (BREWSTER et al., 1992; BREWSTER & ANNAN, 1994). They found that it was possible to observe the movement of a plume of chlorinated hydrocarbons in a sandy aquifer. Perchloroethylene (770 L) was injected into the ground in the center of a 9 m × 9 m area surrounded by a sheet steel wall driven down to the clay layer at a depth of 3.3 m to prevent uncontrolled entry into the surrounding groundwater. A Pulse-Ekko-IV system (200 MHz) was used before and 16 and 917 hours after injection with a 1-m profile spacing and 5 cm spacing between measurement points along a profile. Radargrams along a profile through the center of the test area are shown in Fig. 4.5-18. In the radargram before injection (top), the top of the clay layer appears as a high-energy reflection at about 110 ns. Weaker reflections within the aquifer are caused by laminations in the sand. The inclined reflections are diffractions from the steel walls. After 16 hours (middle), reflections at about 1 and 2 m depth can be recognized. These correlate with elevated PCE concentration. The plume front is already differentiated at this depth but has not spread much laterally. After 917 hours (bottom), strong reflections in the center above the clay layer can be seen, showing that the front has reached the clay layer and is spreading laterally. In another experiment (GREENHOUSE et al., 1993), the “Borden Experiment”, the contaminant front was indicated by “bright spots” (local, elevated amplitudes caused by abrupt changes in reflection properties) in the radargrams. More studies need to be made to determine the practical value of such measurements. A possible application would be the monitoring of sanitation projects.
Environmental Geology, 4 Geophysics
RSW RT D PCE
315
Reflection from sand interbeds Reflection on the top of the clay layer Diffraction Reflection due to high PCE content
Fig. 4.5-18: Radargram from the dense nonaqueous-phase liquids (DNAPL) experiment of BREWSTER et al. (1992), top: before the experiment; middle: 16 hours after the beginning of the percolation of perchloroethylene (PCE) into the ground; bottom: 38 days later, modified after BREWSTER et al. (1992)
316
4.5 Ground Penetrating Radar
4.5.10 Special Applications and New Developments DIETER EISENBURGER
Radar surveys in underground workings There is a need to deposit dangerous materials safely in underground repositories. Because technical barriers only provide a limited degree of security, it is important to find natural geological barriers which are effective and safe in the long term. GPR surveys can be conducted to investigate suitable host rocks. The use of electromagnetic waves for surveying directly in underground workings goes back over 30 years. GPR is a non-destructive method and logistically simpler and more economical to carry out than other exploration methods used for the geological investigation of repositories. Underground GPR surveying methods provide spatial information on the location of structures and heterogeneities by using directional receiving antennas. Such GPR surveys are carried out along profiles in drifts or boreholes. The difficulty with GPR reflection surveys is the extraction of spatial information from a linear antennas array moved along a straight profile. If it is assumed that a reflection surface is in an optically favorable position with respect to the survey profile, then it is possible to calculate the distance to a reflection point (Pn) on the reflecting surface for each survey point if the traveltime and velocity are known (Fig. 4.5-19). The exact position within the reflection plane 13 and the angle O is determined by migration 14 along the profile. However, the angle and distance are not sufficient to determine the precise position of a point in space because a second angle D is required. This is determined from measurements taken along a cross-section perpendicular to the drift axis or by using direction-receiving antennas. For the migration process the wave front method is used because it is able to keep the radial angle D from the survey in this process. This is a robust method which has the advantage of still being applicable even if the distance between survey points does not fulfill the Nyquist condition 15 . If the precise position of the associated reflection points on a reflection surface has been determined, it is possible to construct an element surface for each of these points. If the
13 The reflection plane is defined by the transmitter/receiver position and the reflection
point. 14 Migration is a process by which events on a radargram are mapped to their true spatial
position. It requires a knowledge of the velocity distribution along the raypath. 15 The Nyquist condition is fulfilled when there are more than two samples per cycle
(Nyquist rate) for the highest frequency of the spatial waveform.
Environmental Geology, 4 Geophysics
317
Fig. 4.5-19: Determination of the location of a reflection point: principle
element surfaces of all points are joined, a band in space is generated, visualizing the spatial position of the reflecting surface. GPR surveys in drifts are carried out, if possible, on four profiles: on the floor of the drift, on the two walls and on the roof (Fig. 4.5-20).
Fig. 4.5-20: Configuration of GPR measurements in a drift
It is recommended that this survey is carried out immediately after cutting the drift to prevent the influence of subsequent installations such as cables, pipes, equipment, etc., influencing the survey. The focusing effect of the denser medium (in this case, the surrounding rock) means that reflections from the direction of the profile (floor, wall, roof) are preferentially received. The azimuthal direction from which the individual reflections are received are
318
4.5 Ground Penetrating Radar
more accurately determined by making measurements 360q around a drift perpendicular to the drift axis (Fig. 4.5-20). The azimuthal direction can also be determined using a direction finder (Fig. 4.5-21). The Adcook antenna (Fig. 4.5-21a) is a minimum direction finder; the Rohde & Schwarz antenna (Fig. 4.5-21b) is a maximum direction finder. A radargram is prepared for each profile (examples of radargrams of wall profiles are shown in Fig. 4.5-22) and the individual reflections are picked and compiled to delineate reflectors, which are depicted as a band in space (Fig. 4.5-23). Geological considerations can be used in some cases to identify reflectors as belonging to the same reflecting interface and to construct this reflecting interface by combination and interpolation of the reflectors.
Fig. 4.5-21: GPR surveying equipment with direction finding equipment
Environmental Geology, 4 Geophysics
319
Fig. 4.5-22: Radargrams from the north and south wall profiles along a directional drift
320
4.5 Ground Penetrating Radar
Fig. 4.5-23: Spatial evaluation of the reflectors
GPR single borehole logging GPR borehole logs are carried out in boreholes drilled underground as well as those drilled from the surface. The GPR borehole logging system described here is based on the pulsed radar method described above. It consists of a transmitter with a dipole as the transmission antenna and a receiver with a direction-sensitive receiving antenna, plus the necessary digitizing and recording unit (EISENBURGER et al., 1993). The receiving antenna consists of two orthogonal loop antennas which can be connected to form a dipole. The pulse-generating transmission electronics is built into the transmission antenna. The use of direction-sensitive receiving antennas allows to determine the spatial position of reflection interfaces from a single borehole. Logging is carried out with the tool held at discrete points. The signals received by the two loop antennas, as well as by the dipole formed by them are recorded and used to determine the direction from which a reflection is received using Equation (4.5.13). The dipole signal is used to resolve the ambiguity of the angle derived from the two loop antennas. The angle determined remains stable for the whole time a reflection signal is received (Fig. 4.5-25) and it is a gage for good quality reflections. tan D 1,2
1 2¦ x i y i
ª 2 2 « ¦ xi ¦ y i r ¬
¦ xi2 ¦ y i2
2
2º 4 ¦ x i y i » (4.5.13) ¼
Environmental Geology, 4 Geophysics
321
Several amplitudes are taken from a wavelet recorded with loop 1 (xi) and loop 2 (yi). In addition to the calculated angle (Į) from the loop signals, the precise position of the tool needs to be known. For this reason, the measurement also includes the depth in the borehole and the distortion of the tool around the roll angle (ȕ). The reference position for the roll angle in vertical boreholes is magnetic north, and in the case of horizontal boreholes, the vertical-up position. The absolute direction from which a reflection is received can be calculated from the angle D using (4.5.13) and the roll angle E (Fig. 4.5-24). The method described above can be used to determine the location of the reflectors from the traveltime and the calculated receiving direction. Figures 4.5-26 and 4.5-27 show an example of a radar log using a directional antenna in a vertical borehole.
Fig. 4.5-24: Principle of determination the azimuthal direction of reflections received in boreholes
322
4.5 Ground Penetrating Radar
Fig. 4.5-25: Example of the determination of the azimuthal direction from which a reflection is received
Environmental Geology, 4 Geophysics
Fig. 4.5-26: Radargram of a vertical borehole log with the marked reflectors
323
324
4.5 Ground Penetrating Radar
Fig. 4.5-27: Perspective diagram showing reflectors determined using a direction-sensitive GPR borehole log
Environmental Geology, 4 Geophysics
325
The reflectors determined in this way represent reflection horizons which usually correspond to geological boundaries. This method allows to obtain spatial information on the geological structure of the rocks surrounding a single borehole. This information is used to create a three-dimensional geological model and for the much more precise and better characterization of geological barriers than possible with other methods. All of this information increases the safety and efficiency of the planning of underground repositories. In already existing repositories, e.g., caverns for the storage of hydrocarbons, this borehole logging method can help to acquire additional information on the rock around the caverns – information which is a very valuable supplement on the often inadequately investigated geology, and which, therefore, makes a contribution to improve the economic efficiency and safer operation of a storage cavern.
Helicopter-borne surveys with stepped-frequency radar (SFR) In the past, helicopter-borne GPR systems have been restricted to the use of pulse-radar technology to determine the thickness of polar glaciers. Because of its limited resolution, this method is only of limited suitability for the geological investigation of shallow structures. Stepped-frequency radar (SFR) is a better alternative radar method for shallow geological investigations. Figure 4.5-28 shows how classic pulse-radar works. Transmission pulses (TX) propagate from the transmission antenna and are received by the receiving antenna after a time delay corresponding to the distance to the target.
Fig. 4.5-28: Scheme of pulse radar, showing the transmission pulse (TX) and the receiving pulse (RX)
In Fig. 4.5-28, the upper received signal (RX) represents an idealized situation. The received signal shown in the lower part of the diagram shows the typically distorted measured pulse. The vertical resolution depends on the transmission pulse width. Therefore, a system with improved resolution requires a reduced pulse width. Because the transmitted energy corresponds to the “pulse area” (i.e., amplitude of the pulse integrated over the pulse width), a reduction of pulse width requires an enlargement of transmitted peak power in order to maintain the pulse energy decisive for the penetration depth. Limits to
326
4.5 Ground Penetrating Radar
the achievable vertical resolution arise from technical limits to transmission strength, regulatory limits, or the technical inability to reduce the pulse width. In contrast to classical pulse radar systems, SFR systems operate with amplitude-continuous radar signals (IIZUKA et al., 1984). The signal bandwidth being required for the desired radar resolution is generated sequentially instead of providing the instantaneous complete spectrum of the pulse radar case. Figure 4.5-29 illustrates the basic SFR modulation scheme. The radar transmitter sequentially provides signals stepping through the desired frequency range. Depending on the application this could be done, for instance, in linear steps as depicted in the figure. In the receiver section, both phase and amplitude measurement of the echo signal is performed. The results are fed into a data processing, the well-known IFFT (Inverse Fast Fourier Transform), for example. The output is a pulse that can be compared to that of a pulse radar. It contains the range information of the target, while its pulse width, the radar range resolution, is related to the bandwidth of the applied radar spectrum. Subsequent Measurement of Echoes at Single Frequency Lines f
t
A
A IFFT
Compiled Receive Spectrum
f
Synthesised Pulse
t
Fig. 4.5-29: Scheme of stepped-frequency radar (SFR), calculation of a synthetic pulse from several received echo lines
Environmental Geology, 4 Geophysics
327
The advantages of the SFR can briefly be summarized as follows: low instantaneous bandwidth, high sensitivity, high penetration depth, low sensitivity to radio frequency - interferences, low power consumption, high resolution with respect to measurement frequency, low output data rate (saves memory, allows for high dynamic range AD-converters), reduced wideband antenna problems. Most of the advantages are directly related to the continuous wave operation and the low instantaneous bandwidth of the SFR. The second key factor resulting from the sequential operation principle, is the unique possibility for powerful instrument calibration in both, the frequency as well as in the time domain. The influence of the overall calibration is finally reflected by the achieved radar range resolution performance that is close to theory. This radar method produces better resolution and penetration depths by controlling the range of the frequencies used (Fig. 4.5-30). And because the stepped-frequency technique only requires very low power of the transmitter, it is suitable for surveys where the potential interference with other equipment is to be avoided. A helicopter-borne SFR system (Fig. 4.5-31) makes it possible to quickly investigate large areas and can also be used to fly over dangerous or poorly accessible ground.
Fig. 4.5-30: Example of a stepped-frequency radar line
328
Fig. 4.5-31: SFR helicopter system in operation
4.5 Ground Penetrating Radar
329
Environmental Geology, 4 Geophysics
Physical parameters and units Parameters
Symbols
Units
electric field strength power frequency
E Pt, Pr f
V m-1 W, m2 kg s-3 Hz
angular frequency
Z = 2Sf
s-1, rad s-1
4S10 7 V s A-1 m-1
induction constant of a vacuum P0
8.854 10-12 A s V-1 m-1
velocity of light in a vacuum
H0 c0
Boltzmann constant
k
phase velocity
v
1.380510-23A s K-1 m s-1, m ns-1
permittivity
H*
H c i H cc
A s V-1 m-1
relative permittivity
Hr
H /H0
dimensionless
characteristic impedance (electrical impedance)
Z*
propagation constant
J
absorption coefficient
Dc
dB m-1
electrical resistivity
:m
electrical conductivity
U V
tangent of the loss angle G
tan G
permittivity of a vacuum
2.998 108 m s-1= 0.2998 m ns-1
:
D iE
m-1
S m-1, mS m-1
H cc Hc
V dimensionless ZH c
References and further reading AHRENS, T. J. (Ed.) (1995): A handbook of physical constants. Am. Geophys. Union, Washington, 3 vols. ANNAN, A. P., WALLER, W. M., STRANGWAY, D. W., ROSSITER, J. R., REDMAN, J. D. & WATTS, R. D. (1975): The electromagnetic response of a low loss, 2-layer, dielectric earth for horizontal electric dipole excitation. Geophysics, 40, 285-298. BALANIS, C. A. (1996): Antenna theory: analysis and design. Wiley. BAÑOS, A. (1966): Dipole radiation in the presence of a conducting halfspace. Pergamon Press, New York. BEBLO, M. (1982): Elektrische Eigenschaften. In: ANGENHEISTER, G. (Ed.): LANDOLT BÖRNSTEIN: Zahlenwerte und Funktionen aus Naturwissenschaften und Technik. Neue Serie V, 1b, 254-261, Springer, Berlin.
330
4.5 Ground Penetrating Radar
BLEIL, U. & PETERSEN, N. (1982): Magnetic properties: in Landolt-Bornstein numerical data and functional relationships in science and technology: group V, geophysics and space research, 1b, physical properties of rocks. ANGENHEISTER, G. (Ed.), Springer, Berlin, 308-432. BLINDOW, N. (1986): Bestimmung der Mächtigkeit und des inneren Aufbaus von Schelfeis und temperierten Gletschern mit einem hochauflösenden elektromagnetischen Reflexionsverfahren. Dissertation, Institut für Geophysik, Westfälische WilhelmsUniversität Münster. BLINDOW, N., ERGENZINGER, P., PAHLS, H., SCHOLZ, H. & THYSSEN, F. (1987): Continuous profiling of subsurface structures and groundwater surface by EMR methods in Southern Egypt. Berliner Geowiss. Abh. (A) 75.2, 575-627. BOHIDAR, R. N. & HERMANCE, J. F. (2002): The GPR refraction method. Geophysics, 67, 1474-1485. BREKOVSKIKH, L. M. (1980): Waves in layered media, 2nd edn. Academic Press, New York. BREWSTER, M. L., ANNAN A. P. & REDMAN, J. D. (1992): GPR Monitoring of DNAPL migration in a sandy aquifer. In: Fourth international conference on ground penetrating radar. Geological Survey of Finland, Special Paper 16, 185-190. BREWSTER, M. L. & ANNAN, A. P. (1994): Ground-penetrating radar monitoring of a controlled DNAPL release: 200 MHz radar. Geophysics, 59, 1211-1221. BREWSTER, M. L., ANNAN, A. P., GREENHOUSE, J. P., KUEPER, B. H., OLHOEFT, G. R., REDMAN, J. D. & SANDER, K. A. (1995): Observed migration of a controlled DNAPL release by geophysical methods: Ground Water. 33, 977-987. BRISTOW, C .S. & JOL, H. M. (2003): Ground penetrating radar in sediments. Geological Society Publication 211, London. CAI, J. & MCMECHAN, G. A. (1995): Ray-based synthesis of bistatic ground-penetrating radar profiles. Geophysics, 60, 87-96. CARCIONE, J. M. & SERIANI, G. (2000): An electromagnetic modelling tool for the detection of hydrocarbons in the subsoil. Geophys. Prosp., 48, 231-256. CARMICHAEL, R. S. (Ed.) (1982): Handbook of physical properties of rocks. CRC Press, Boca Raton, 3 vols. CLOUGH, J. W. (1976): Electromagnetic lateral waves observed by earth sounding radars. Geophysics, 41, 1126-1128. CONYERS, L. B. & Goodman, D. (1997): Ground-penetrating radar: an introduction for archaeologists. Altimira. DANIELS, D. J., GUNTON, D. J. & SCOTT, H. F. (1988): Introduction to subsurface radar. IEE Proceedings F, 135 (F,4), 278-320. DANIELS, J. J. (1989): Fundamentals of ground penetrating radar. Proceedings of the Symposium on the Application of Geophysics to Engineering and Environmental Problems, SAGEEP 89, Golden, Colorado, 62-142.
331
Environmental Geology, 4 Geophysics
DANIELS, J. J. (1996): Surface Penetrating Radar. The Institution of Electrical Engineers, London. DANIELS, J. J. (2004): Ground Penetrating Radar – 2nd edn. The Institution of Electrical Engineers, London. DANIELS, J. J. (1989): Fundamentals of ground penetrating radar. Proceedings of the Symposium on the Application of Geophysics to Engineering an Environmental Problems, SAGEEP 89, Golden, Colorado, 62-142. DAVIS, J. L. & ANNAN, A. P. (1989): Ground penetrating radar for high-resolution mapping of soil and rock stratigraphy. Geophys. Prosp., 37, 531-551. DOOLITTLE, J. A. (1982): Characterizing soil map units with the ground-penetrating radar. Soil Surv. Horizons, 23, 4, 3-10. DOOLITTLE, J. A. (1983): Investigating Histosols with the ground-penetrating radar. Soil Surv. Horizons, 24, 3, 23-28. DOUGLAS, D. G., BURNS, A. A., RINO, CH. L. MARESCA, J. W. (1992): A study to determine the feasibility of using a ground-penetrating radar for more effective remediation of subsurface contamination. Risk Reduction Engineering Laboratory Office of Research and Development U. S. Environmental Protection Agency, Cincinnati, Ohio, Report EPA/600/R-92/089. EISENBURGER, D., SENDER, F. & THIERBACH, R. (1993): Borehole Radar- An Efficient Geophysical Tool to Aid in the Planing of Salt caverns and Mines. Seventh Symposium on Salt, I, 279-284, Elsevier, Amsterdam. FORKMANN, B. & PETZOLD, H. (1989): Prinzip und Anwendung des Gesteinsradars zur Erkundung des Nahbereichs. Freiberger Forschungshefte, C 432, Dt. Verlag für Grundstoffindustrie, Leipzig. FORKMANN, B. & PETZOLD, H. (1991): Gesteinsradar - Prinzip und Anwendungsmöglichkeiten. Z. angew. Geol., 37, 25-30. GEVANTMAN, L. H. (Ed.) (1981): Physical properties data for rock salt. National Bureau of Standards Monograph 167. GOODMAN, D. (1994): Ground-penetrating archaeology. Geophysics, 59, 224-232.
radar
simulation
in
engineering
and
GPR 1994, Proceedings of the Fifth International Conference on Ground Penetrating Radar; 12-16 June, 1994, Kitchener, Ontario Canada. GPR 1996, Proceedings of the Sixth International Conference on Ground Penetrating Radar; September 30- October 3, 1996, Tohoku Japan. GPR 1998, Proceedings of the Seventh International Conference on Ground Penetrating Radar; 27-30 May, 1998, Lawrence, Kansas USA. GPR 2000, Proceedings of the Eighth International Conference on Ground Penetrating Radar; 23-26 May, 2000, Gold Coast, Australia. GPR 2002, Proceedings of the Ninth International Conference on Ground Penetrating Radar; April 29- May 2, 2002, Santa Barbara, California, USA.
332
4.5 Ground Penetrating Radar
GPR 2004, Proceedings of the Tenth International Conference on Ground Penetrating Radar; 21 – 24 June, 2004, Delft, The Netherlands. GPR 2006, Proceedings of the Eleventh International Conference on Ground Penetrating Radar; June 19-22, 2006, Columbus, Ohio, USA. GREENHOUSE, J., BREWSTER, M., SCHNEIDER, G., REDMANN, D., ANNAN, P., OLHOEFT, G., LUCIUS, J., SANDER, K. & MAZELLA, A. (1993): Geophysics and solvents: the Borden experiment. The Leading Edge, 261-267. HALLEUX, L., FELLER, P., MONJOIE, A. & PISSART, R. (1992): Ground penetrating and borehole radar surveys in the Borth salt mine (FRG). In: Fourth International Conference on Ground Penetrating Radar. Geological Survey of Finland, Special Paper 16, 317-321. HALLEUX, L. & RICHTER, T. (1994): Radar tomography for shallow engineering geological investigations. Poster presentation on the Fifth International Conference on Ground Penetrating Radar; June 12–16, 1994, Kitchener, Ontario, Canada. HANSEN, V. W. (1989): Numerical solution of antennas in layered media. Wiley. V. HIPPEL,
A. R. (1954): Dielectrics and waves. M. I. T. Press, Cambridge, Massachusetts.
HUGGENBERGER, P., MEIER, E. & PUGIN, A. (1994): Ground-probing radar as a tool for heterogeneity estimation in gravel deposits: Advances in data processing and facies analysis. Applied Geophysics, 31, 171-184. IIZUKA, K., FREUNDORFER, A. P., WU, K. H., MORI, H., OGURA, H. & NGUYEN, V.-K. (1984): Step-frequency radar. J. Appl. Phys., 56, 9, 2572-2583. JANSCHEK, H., MAURITSCH, H., RÖSLER, R. & STEINHAUSER, P. (1985): Hochfrequenzmethoden. In: MILITZER, H. & WEBER, F. (Eds.): Angewandte Geophysik, 2, Geoelektrik-Geothermik-Radiometrie-Aerogeophysik. Springer Wien, Akademie Berlin, 151-173. JOHARI, G. P. & CHARETTE, P. A. (1975): The permittivity and attenuation in polycrystalline and single-crystal ice Ih at 35 and 60 MHz. J. Glac., 14, 293-303. JONES, F. H. M., NAROD, B. B. & CLARKE, G. K. C. (1989): Design and operation of a portable impulse radar. J. Glac., 35, 143-147. KING, R. W. P. (1980): Antennas in matter: fundamentals, theory, and applications: M.I.T. Press, Cambridge, Massachusetts. KRAUS, J. D. & Marhefka, R. J. (1988): Antennas, 2nd edn.: McGraw-Hill, New York. KRAUS, J. D. (1991): Electromagnetics, 4th edn.: McGraw-Hill, New York. LAMBOT, S. & GORITTI, A. (Eds.) (2007): Special Issue on Ground Penetrating Radar. Near Surface Geophysics, 5, 1, 5-82. LAMPE, B., HOLLIGER, K. & GREEN, A. G. (2003): A finite difference time-domain simulation tool for ground-penetrating radar antennas. Geophysics, 68, 971-987. LEPAROUX, D., GIBERT, D. & COTE, P. (2001): Adaptation of prestack migration to multioffset ground penetrating radar (GPR) data. Geophys. Prosp., 49, 374-386.
Environmental Geology, 4 Geophysics
333
LUCIUS, J. E., OLHOEFT, G. R., HILL, P. L. & DUKE, S. K. (1992): Properties and hazards of 108 selected substances - 1992 edn: U.S. Geological Survey Open-File Report 92-527. MCMECHAN, G. A., LOUCKS, R. G., MESCHER, P. & ZENG, X. (2002): Characterization of a coalesced, collapsed paleocave reservoir analog using GPR and well-core data. Geophysics, 67, 1148-1158. MEHLHORN, H., RICHTER, TH. BAND, S. (1988): Beitrag zu theoretischen Aspekten der Anwendung des Radarverfahrens zur Ortung von punkt- oder flächenartigen Zielen. Neue Bergbautechnik, 18, 176-178. MEYERS, R. A., SMITH, D. G., JOL, H. M. & PETERSON, C. D. (1996): Evidence for eight great earthquake-subsidence events detected with ground-penetrating radar. Willapa Barrier, Washington. Geology, 24, 99-102. MILITZER, H. & WEBER, F. (1987): Angewandte Geophysik, 3, Seismik. Springer, Berlin, Akademie Berlin. MILLER, E. K. (Ed.) (1986): Time-domain measurements in electromagnetics. Van Nostrand, New York. MOREY, R. M. (1998): Ground penetrating radar for evaluating subsurface conditions for transportation facilities. NAS/NRC/TRB NCHRP Synthesis Report 255. MUNDRY, E. (1991): Numerische Modelluntersuchungen zur Reflexion hochfrequenter elektromagnetischer Wellen im Salzgestein. Geol. Jb., E 48, Hannover, 259-282. NOON, D. A., LONGSTAFF, D. YELF, R. J. (1994): Advances in the development of step frequency ground penetrating radar. Proceedings of the Fifth International Conference on Ground Penetrating Radar. Kitchener, Ontario, June 12-16, 117-131. OLHOEFT, G. R. (1988): Interpretation of hole-to-hole radar measurements. In: Proceedings of the Third Technical Symposium on Tunnel Detection, January 12-15, 1988, Golden, CO, 616-629. OLHOEFT, G. R. (1992): Geophysical detection of hydrocarbon and organic chemical contamination. In: BELL, R. S., (Ed.): Proceedings on Application of Geophysics to Engineering, and Environmental Problems, Oakbrook, IL: Society of Engineering and Mining Exploration Geophysics, Golden, CO, 587-595. OLHOEFT, G. R. CAPRON, D. E. (1994): Petrophysical causes of electromagnetic dispersion. Proceedings of the Fifth International Conference on Ground Penetrating Radar. Kitchener, Ontario, June 12-16, 145-152. OLHOEFT, G. R., POWERS, M. H. & CAPRON, D. E. (1994): Buried object detection with ground penetrating radar. In: Proc. of Unexploded Ordnance (UXO) detection and range remediation conference, Golden, CO, May 17-19, 1994, 207-233. OLHOEFT, G. R. (1998): Electrical, magnetic and geometric properties that determine ground penetrating radar performance. In: Proc. of GPR'98, 7th Int’l. Conf. On Ground Penetrating Radar, May 27-30, 1998, The Univ. of Kansas, Lawrence, KS, USA, 177-182. OLHOEFT, G. R. (2000): Maximizing the information return from ground penetrating radar. J. Applied Geophys., 43, 175-187.
334
4.5 Ground Penetrating Radar
OLHOEFT, G. R. (2004): www.g-p-r.com. Webpage on Ground Penetrating Radar with a GPR Tutorial, a Bibliography and links to GPR Manufactures. PFEIFFER, W. (1976): Impulstechnik. Hanser, München. PIPAN, M., FORTE, E., GUANGYOU, F. & FINETTI, I. (2003): High resolution imaging and joint characterization in limestone. Near Surface Geophysics, 1, 39-55. RADEZEVICIUS S. J., CHEN C. C., PETERS L. & DANIELS J. J. (2003): Near-field dipole radiation dynamics through FDTD modelling. Journal of Applied Geophysics, 52, 75-91. REYNOLDS, J. M. (1997): An Introduction to Applied and Environmental Geophysics. John Wiley & Sons Ltd., Chichester. ROTH, F., VAN GENDEREN, P. & VERHAEGEN, M. (2004): Radar Scattering Models for the Identification of Buried Low-Metal Content Landmines. Proceedings of the Tenth International Conference on Ground Penetrating Radar; 21–24 June, 2004, Delft, The Netherlands. ROTHAMMEL, K. (1991): Antennenbuch. 10th edn., Franck-Kosmos, Stuttgart. SKOLNIK, M. I. (1970): Introduction to radar systems. McGraw-Hill, New York. SLOB, E. & YAROVOY, A. (Eds.) (2006): Special Issue on Ground Penetrating Radar. Near Surface Geophysics, 4, 1, 5-75. STERN, W., (1929): Versuch einer elektrodynamischen Dickenmessung von Gletschereis. Gerl. Beitr. zur Geophysik, 23, 292-333. STERN, W., (1930): Über Grundlagen, Methodik und bisherige Ergebnisse elektrodynamischer Dickenmessung von Gletschereis. Z. Gletscherkunde, 15, 24-42. SZERBIAK, R. B., MCMECHAN, G. A., CORBEANU, R., FORSTER, C. & SNELGROVE, S. H. (2001): 3-D characterization of a clastic reservoir analog: From 3-D GPR data to a 3-D fluid permeability model. Geophysics, 66, 1026-1037. TAYLOR, B. N. (1995): Guide for the use of the international system of units (SI). NIST Spec. Publ. 811, 1995 ed., USGPO, Washington, DC. (http://www.nist.gov). THIERBACH, R. (1974): Electromagnetic reflections in salt deposits. J. Geophys., 40, 633-637. THYSSEN, F. (1985): Erkundung oberflächennaher Strukturen und Eigenschaften mit dem elektromagnetischen Reflexionsverfahren. In: HEITFELD, K.-H. (Ed.): Ingenieurgeologische Probleme im Grenzbereich zwischen Locker- und Festgesteinen. Springer, Berlin, 597-609. TRONICKE, J., DIETRICH, P., WAHLIG, U. & APPEL, E. (2002): Integrating surface georadar and crosshole radar tomography: A validation experiment in braided stream deposits. Geophysics, 78, 1516-1523. TURNER, G. & SIGGINS, A. F. (1994): Constant Q-attenuation of subsurface radar pulses. Geophysics, 59, 1192-1200.
Environmental Geology, 4 Geophysics
335
TSANG, T., KONG, J. A. & SIMMONS, G. (1973): Interference patterns of a horizontal electric dipole over layered dielectric media. J. Geophys. Res., 78, 3287-3300. TILLARD, S. (1994): Radar experiments in isotropic and anisotropic geological formations (granite and schists). Geophys. Prosp., 42, 615-636. VAN OVERMEEREN, R. A. (1994): Georadar for hydrogeology. First Break, 12, 401-408. WRIGHT, D. L., HODGE, S. M., BRADLEY, J. A., GROVER, T. P. & JACOBEL, R. W. (1990): A digital low-frequency, surface-profiling ice-radar system. J. Glac., 36, 1112-1121. WU, T. T. & KING, R. W. P. (1965): The cylindrical antenna with nonreflecting resistive loading, IEEE Trans. Antennas and Propagation AP-13, 369-373. WYATT, D. E. & TEMPLES, T. J. (1996): Ground-penetrating radar detection of small-scale channels, joints and faults in the unconsolitated sediments of the atlantic coastal plain. Environmental Geology, 27, 219-225. ZENG, X. & MCMECHAN, G. A. (1997): GPR characterization of buried tanks and pipes. Geophysics, 62, 797-806. ZENG, X., MCMECHAN, G. A. & XU, T. (2000): Synthesis of amplitude-versus-offset variations in ground-penetrating radar data. Geophysics, 65, 113–125.
Environmental Geology, 4 Geophysics
337
4.6 Seismic Methods ANDREAS SCHUCK & GERHARD LANGE
4.6.1 Principle of the Methods The basic principle of all seismic methods is the controlled generation of elastic waves by a seismic source in order to obtain an image of the subsurface. Seismic waves are pulses of strain energy that propagate in solids and fluids. Seismic energy sources, whether at the Earth’s surface or in shallow boreholes, produce wave types known as: x body waves, where the energy transport is in all directions, and x surface waves, where the energy travels along or near to the surface. Two main criteria distinguish these wave types from each other - the propagation zones and the direction of ground movement relative to the propagation direction. Of prime interest in shallow seismics are the two types of body waves: x P-waves (primary, longitudinal or compressional waves) with particle motion parallel to the direction of propagation, and x S-waves (secondary, shear or transverse waves) with particle motion perpendicular to the direction of propagation – when particle motion is in the vertical plane they are referred to as SV-waves, and SH-waves when the particle motion is in the horizontal plane. Surface waves, often considered to be a source of noise, contain valuable information about material properties of the shallow ground. Their use is on the increase in engineering studies. The velocity of seismic waves is the most fundamental parameter in seismic methods. It depends on the elastic properties as well as bulk densities of the media and varies with mineral content, lithology, porosity, pore fluid saturation and degree of compaction. P-waves have principally a higher velocity than S-waves. S-waves cannot propagate in fluids because fluids do not support shear stress. During their propagation within the subsurface seismic waves are reflected, refracted or diffracted when elastic contrasts occur at boundaries between layers and rock masses of different rock properties (seismic velocities and/or bulk densities) or at man-made obstacles. The recording of seismic waves returning from the subsurface to the surface allows drawing conclusions on structures and lithological composition of the subsurface. By measuring the
338
4.6 Seismic Methods
traveltimes of seismic waves and determining their material specific velocities, a geological model of the subsurface can be constructed. In a seismic survey elastic waves are generated by different energy sources (e.g., explosives, vibrator, weight-drop, sledgehammer). The seismic response is simultaneously recorded by a number of receivers (geophones, seismometers). These are positioned along straight profile lines (2-D seismics) or over an area in 3-D seismic surveying and connected to a seismograph. The signals of individual geophones or groups of geophones are recorded by a seismograph, processed and displayed in seismic sections to image the subsurface structure. The classical seismic methods are refraction seismics (Section 4.6.5) and reflection seismics (Section 4.6.6). Applications of seismic methods on inshore waters or in boreholes are also standard practice (Fig. 4.6-1). In refraction seismics “head waves” (or Mintrop waves) are used, which arise at the interface between two layers when refraction at the critical angle occurs. The critical angle is that angle of ray incidence for which the refracted ray moves along the contact surface (Fig. 4.6-12). This requires an increase in velocity with depth. If a layer of lower seismic velocity underlies a layer of higher velocity (velocity reversal), no refraction at the critical angle occurs and the layer cannot be detected (hidden layer). Such conditions may cause ambiguity in the interpretation. In the refraction analysis the traveltimes of first arrivals on recorded seismograms are commonly used. The first arrivals have to be identified, picked and presented in traveltime curves (plots of arrival times vs. source-receiver distance). Main aim of refraction surveys is to determine the depths to layers, the refractor topography (dip of layers) and layer velocities. To extract this information from traveltime curves, some methods have been developed (e.g., HAGEDOORN, 1959; PALMER, 1981; VAN OVERMEEREN, 2001) and user-friendly software packages, running on PCs are available to assist in the data interpretation. Reflection seismics is the most important method to prospect for oil and gas at greater depths. During the 1980s the importance of the seismic reflection technique in shallow investigations, typically less than 50 m, has increased (e.g., DAVIES et al., 1992; HILL, 1992b; KING, 1992). It must however, be pointed out that the application of reflection seismics to shallow investigations, is not merely a case of transferring the seismic technology used so successfully in prospecting for hydrocarbons. The unprocessed field records are mostly characterized by poor signal-to-noise ratios that result in some special problems for shallow seismic surveys. Because of the short traveltimes, source generated noise, surface waves, first arrivals of direct waves and refracted waves dominate and genuine reflections are masked (Fig.4.6-15).
Environmental Geology, 4 Geophysics
339
Fig. 4.6-1: Principles of seismic methods. Dashed lines show raypaths of seismic waves from the source to receivers; v0, v1, v2, v3 are seismic velocities in the respective layers. The returning signal is recorded by geophones at distances much greater (refraction seismics), at distances less than the investigation depth (reflection seismics) and at distances mostly equal to the investigation depth (seismic surface waves). In VSP and crosshole seismics the geophones (or hydrophones) are located in boreholes and the source is positioned at the surface (VSP) or in boreholes (crosshole seismics).
340
4.6 Seismic Methods
Thus, both acquisition and standard processing as performed in conventional seismic reflection methods cannot be transferred by simply using a “scaling down” approach. Some possible sources of misinterpretation resulting from the complex wavefields appearing in shallow reflection seismic records are described by STEEPLES & MILLER (1998). The surface wave method is based on analyzing the variation of seismic velocities with frequency (dispersion). Dispersion of surface waves occurs when near surface velocity layering is present (Section 4.6.3.4). Since the late 1990s the method has been successfully used in many applications, as for example underground cavity detection and the delineation of abandoned waste sites (e.g., PARK et al., 1999; HAEGEMAN & VAN IMPE, 1999; LEPAROUX et al., 2000; SHTIVELMAN, 2002).
4.6.2 Applications The main applications with respect to waste disposal and contaminated sites, civil engineering and foundation studies, and in the search for groundwater can be summarized as follows: x investigation of regional and local geological structures, x detection of faults, joints and other zones of weakness, x identification of shallow stratigraphic sequences, x determination of depth to bedrock and relief of bedrock surface, x investigation of layer boundaries, in particular with respect to groundwaterbearing formations, x distribution and thickness of weathering layer and of erosion channels, x delineation of areas of different lithology and facies, x mapping the topography of the base of disposal sites, x investigations below sealed surfaces, x investigation of ground deformations caused by mining and subrosion, x determination of elastic parameters (Poisson’s ratio, shear modulus, elastic bulk modulus, Lamé parameter, etc. – see Table 4.6-1), especially rock competence for geotechnical application, x localization of manmade subsurface structures (e.g., buildings, tanks, foundations), and x localization of cavities and voids.
Environmental Geology, 4 Geophysics
341
4.6.3 Fundamentals 4.6.3.1 Propogation of Elastic Waves During their propagation within the ground elastic waves disturb the medium causing particle motion as well as volume and shape changes. A mathematical expression of particle displacement \ and wave velocity v as functions of space (x, y, z) and time t is the wave equation. The wave equation is based on the theory of elastic continua which relates deformation (strain) and stress. Equation (4.6.1) describes the wave propagation in a homogeneous isotropic elastic medium, where \ can be a vector or a scalar, e.g., a component u of the particle displacement: 2\
w 2\ w 2\ w 2\ 2 2 wx wy wz 2
§ 1 · w 2\ ¨ 2¸ 2 . © v ¹ wt
(4.6.1)
The simplest solution of the wave Equation (4.6.1) is that of a planar wave with harmonic ground motion
\ t
A cos I Z t
(4.6.2)
with A being the amplitude, Z the angular frequency (2Sf) and I the phase of the wave. Adding an additional phase term k z delivers a generalized form of Equation (4.6.2) in the z direction of wave propagation (1-dimensional case):
\ t
A cos I Z t k z ,
(4.6.3)
where k is the wavenumber (k = 2S /O) with O being the wavelength, and z the distance along the wave propagation. However, planar harmonic waves are idealized waveforms which can be used as approximation only in a very large distance to the source. Solutions of the wave equation for several types of wavefront geometries (cylindrical waves, spherical waves, etc.) can be found, for example, in TELFORD et al. (1990). For very small deformations of a homogeneous isotropic elastic medium strain is directly proportional to stress (Hooke’s law) - an assumption which is valid for wave propagation in many geological materials. The relations between stress (force per unit area) and strain (change of volume and shape) are described by the general elasticity tensor, which contains up to 21 independent elastic constants for anisotropic media. These elastic constants are expressed in terms of each other, P- and S-wave velocities (vp and vs), and bulk density U (Table 4.6.1 and SHERIFF, 1991). The simplest assumption is seismic wave propagation in an isotropic medium. In this case only two elastic constants, noted as Lamé’s constant O (lambda constant) and µ (shear modulus), exist and the solution of the wave equation yields two different O 2P / U and v s P / U . Both phase velocities phase velocities v p
342
4.6 Seismic Methods
are independent of the direction of wave propagation and identical with the group velocity 1. The wave traveling in an isotropic medium at the higher velocity vp (P-wave) is polarized parallel to the direction of wave propagation. Waves with lower velocity vs (S-wave) can be split into a wave vertically polarized (SV-wave) and a wave horizontally polarized (SH-wave) perpendicular to the direction of propagation. P-waves cause changes of shape and volume, whereas S-waves only cause changes to the shape of the medium through which they are transmitted (Fig. 4.6-2).
Fig. 4.6-2: Particle motion of a compressional wave (left) and a shear wave (right), when passing through an elastic material
1 Phase velocity is the velocity of any phase (peak or trough) of an event correlated on a
seismic record. Group velocity is the velocity with which the seismic energy in a wavetrain travels. Because of dispersion the phase velocity may differ from the group velocity.
Environmental Geology, 4 Geophysics
343
In an inhomogeneous medium both wave types are coupled. Seismic energy, traveling partly as a P-wave and partly as an S-wave, is converted from one to the other when reflection or refraction occurs at oblique incidence on a boundary in the elastic properties (seismic boundary). Since mode conversion is small for low incident angles, converted waves become more prominent for large source-receiver distances (offsets). In the most frequent case of an anisotropic medium (e.g., fine layered ground), the velocities of P-, SV- and SH-waves are different and depend on the direction of wave propagation (HELBIG, 1994). Phase and group velocity are not identical and the P- and Swaves are polarized orthogonal to each other, but the polarization, generally, does not coincide with the direction of wave propagation. For spherical waves radiating from a small source (point source) at or near to the Earth’s surface, wavefronts propagate outward and away from their point of origin (Fig. 4.6-8). Their seismic energy is distributed over a sphere with an area a = 4 S r2 (r being the source distance). This results in a geometrical wave energy decay inversely proportional to the square of the distance (1/r2). The geometrical decay is independent of the wave frequency and called spherical or geometric divergence. A decrease in amplitude during wave propagation is also caused by energy loss due to the non-elastic behavior of real media. This phenomenon, known as absorption or intrinsic attenuation, is caused by viscous energy dissipation, e.g., by movement of fluids in the pore space of sediments. Absorption is strongly dependent on wave frequency. The amplitudes of waves with higher frequencies are attenuated more than those of waves with lower frequencies. Therefore, for low frequencies and short source distances, spherical divergence influences the wave amplitude more than absorption. However, with increasing distance and higher frequencies absorption dominates. In a first order approximation energy loss with increasing distance due to absorption can be regarded as being exponentially. If the medium is characterized by an absorption coefficient (or attenuation factor) D =D (f), the relation between the source energy E0 and the actual energy E at a distance x from the source is E = E0 e-D x. Hence, the decrease in amplitude A or energy E is described in a logarithmic scale and expressed in decibel: decrease = 10 log (E0/E) = 20 log (A0/A) [dB]. 4.6.3.2 Elastic Parameters and Seismic Velocities Elastic parameters For small deformations of an elastic medium the relation between stress and deformation is described by Hooke’s law (Section 4.6.3.1). Depending on the deformation (compressional or shear) the proportionality between stress and strain in the linear range is specified by elastic moduli (Fig. 4.6-3,
344
4.6 Seismic Methods
Table 4.6-1). In addition, a relation exists between elastic moduli and the velocity of seismic waves. Elastic (dynamic) parameters and velocities as derived from non-destructive testing by seismic measurements are used to describe soil mechanical properties in ground engineering, for example for the construction of large buildings. Their determination is standardized and described in recommendations of the American Society for Testing Materials (ASTM), the British Standards Institution and the German Society for NonDestructive Testing (DGZfP). Table 4.6-1: Interrelations between elastic constants, P- and S-wave velocities, and bulk density U for isotropic media
O, µ Lamé’s constant
O
O
shear modulus
µ
µ
bulk modulus
k
O
Young’s modulus
E
P
E, V
v p2 , v s2
EV 1 V 1 2V
U v p2 2v s2
µ
E 2 1 V
U v s2
k
E 3 1 2V
k, µ 2 P 3
k
2 P 3
3O 2P O P
O
Poisson’s ratio
V
2O P
P-wave velocity
v p2
O 2P U
S-wave velocity
v s2
P U
9k P 3k P
3k 2P 6k 2P
k
U P U
4 P 3
E
§ ©
U ¨ v p2
UV p2
3v p2 4v s2 v p2 v s2 v p2
V
4 2· vs 3 ¸¹
2 v s2 v2 2 p2 2 vs
E 1 V
U 1 V 1 2V
v p2
E 2U 1 V
v s2
345
Environmental Geology, 4 Geophysics
a) Poisson’s ratio:
V
b) bulk modulus:
k
c) Young’s modulus:
E
d) shear modulus:
P
'd / d
'l / l pV / 'V s 'l / l
WD
(p hydrostatic pressure; V volume) (s axial stress) (W shear stress)
Fig. 4.6-3: Illustration of elastic deformation of a cylindrical isotropic sample under different stress conditions and mechanical constraints, BERCKHEMER (1990)
Seismic velocities The most important relationship for the joint interpretation of P- and S-wave sections is the velocity-ratio vp/vs. As vp/vs is directly proportional to the traveltime-ratio ts/tp, it can be directly derived from traveltime differences in P- and S-wave sections. For a Poisson’s ratio V = 0.25 the velocity-ratio v p vs 3 . This often holds for consolidated sediments and solid rocks. For shallow unconsolidated and slightly consolidated sediments, the vp/vs-ratio varies between 2 and 12. The simultaneous observation of both P- and Swaves allows drawing conclusions on lateral changes in the lithology, and on the type and degree of fluid saturation (BACHRACH et al., 1998; MAVKO et al., 1998). Whereas P-waves provide information on lithology and fluid content, S-waves contain information about lithology. Table 4.6-2 gives a compilation of velocities and bulk densities of unconsolidated sediments, rocks, and fluids.
346
4.6 Seismic Methods
Table 4.6-2: Seismic velocities and bulk densities of unconsolidated sediments, selected rocks, and fluids, FERTIG (2005) Material clay -sandy clay -loamy clay sand -dry -wet -saturated -coarse -fine grained gravel -wet weathering layer marly limestone marlstone siltstone sandstone -loose -compacted -siliceous quartzite schistose quartzite lime -chalky -fine grained -crystalline marble chalk limestone dolomite gypsum anhydrite halite bituminous coal brown coal air (depending on temperature) petroleum water saltwater (seawater)
Velocity vp [m s-1] 500 - 2800 2000 - 2750 500 - 1900 100 - 2000 100 - 600 200 - 2000 1300 - 1800 1835 1740 180 - 1250 750 - 1250 100 - 500 3200 - 3800 1300 - 4500 1900 - 2000 800 - 4500 1500 - 2500 1800 - 4300 2200 - 2400 5800 5500 3000 - 6000 3560 4680 5500 5100 1800 - 3500 2000 - 6250 2000 - 6250 1500 - 4600 4500 - 6500 4500 - 6500 1600 - 1900 500 - 1800 310 - 360 1035 - 1370 1430 - 1590 1400 - 1600
Velocity vs [m s-1] 110 - 1500
Bulk density U [g cm-3] 1.25 - 2.32
440 - 1080 100 - 500
0.76 - 1.57 1.80 - 2.05 2.33 - 2.80 1.50 - 2.00 1.80 - 2.05 2.03 1.98 1.95 - 2.20 1.95 - 2.20 1.20 - 1.80 2.65 - 2.73
715 - 2250 320 - 2700 575 - 1101 672 - 1023
2.35 - 2.73 2.30 - 2.55 1.80 - 2.40 2.22 - 2.69 2.64 2.65
1800 - 3800 2900 - 3740 750 - 2760 750 - 3600 2250 - 3300 800 - 1140
2.67 2.66 - 2.70 2.00 - 2.57 1.75 - 2.88 1.75 - 2.88 2.31 - 2.33 2.15 - 2.44 2.15 - 2.44 1.25 - 1.84 1.20 - 1.50 1.29·10-3 0.92 - 1.07 0.98 - 1.01 1.01
Environmental Geology, 4 Geophysics
347
4.6.3.3 Reflection, Transmission and Diffraction To describe the seismic wave propagation it is important to know the behavior of seismic waves at an interface. When an advancing planar wavefront2 arrives at a boundary between two layers of different elastic properties, every point at this boundary acts as a source of a secondary spherical wave (Huygens’ principle). These secondary waves overlap and new wavefronts defined by the envelope to all the secondary waves are formed - a reflected wavefront in the upper halfspace and a transmitted (refracted) wavefront in the lower halfspace. For oblique incidence converted waves arise, traveling partly as P-waves and partly as S-waves. The ratio between reflected and transmitted energy and, hence, the ratio of the wave amplitudes, depends on the elastic properties of both layers, and is expressed as their acoustic impedance. The acoustic impedance of a layer is defined by its seismic velocity multiplied by the bulk density. Based on the contrast in acoustic impedances at a seismic boundary, reflection and transmission coefficients can be calculated. These values define that part of energy which is reflected or transmitted. A reflection coefficient of value “1” (theoretically) means that all incident seismic energy is reflected. To calculate the reflection and transmission coefficients at an interface it is necessary to match boundary conditions for the displacement vector u (x, t) and the stress tensor. A simple example to illustrate the calculation of reflection and transmission coefficients is the case when a planar wave incidents normal to an interface between two rigidly connected solid media (Fig. 4.6-4). Keeping the boundary conditions for displacement and stress one can state: The reflection coefficient R (reflectivity) for displacement amplitudes u0 of the incident wave and ur of the reflected wave is: ur u0
R
U2v 2 U1v1 . U2v 2 U1v1
(4.6.4)
The transmission coefficient T for the displacement amplitude uT of the transmitted wave is: uT u0
T
U1v1 U1v1 U2v 2
1 R.
(4.6.5)
2 A wavefront is that surface over which the phase of a wave is the same. At large distances
from the source a wavefront can be considered as planar. As it is simpler to consider wave propagation in terms of rays, analog to the ray-geometric optics, in the following rays and raypaths are considered. A ray is the normal to the wavefront.
348
4.6 Seismic Methods
Fig. 4.6-4: Reflection and transmission coefficients R and T for a planar interface between two half-spaces with the acoustic impedances I1 = U1v1 and I2 = U2v2, respectively. The incident wave I (here shown as a ray normal to the wavefront) has the normalized amplitude of value 1.
Fig. 4.6-5: Ratio of reflected energy to incident energy Er/E0 as a function of the incident angle of a P-wave raypath impinging on the planar interface between medium 1 and 2. In the upper picture the ratio of bulk densities U1/U2 is kept constant and the velocity ratio varies. In the lower picture the velocity ratio remains constant and the density ratio varies. In both cases Poisson’s ratio V = 0.25.
349
Environmental Geology, 4 Geophysics
However, this only holds for normal incidence of a ray. In the general case reflection and transmission coefficients vary with the angle of incidence. Angle-dependent reflection and transmission coefficients for elastic plane waves at a non-slip horizontal boundary are given by the Zoeppritz equations (AKI & RICHARDS, 1980; SHUEY, 1985; www.crewes.org). In Fig. 4.6-5 energy ratios are presented which are proportional to the square of displacement amplitudes. Due to the general boundary conditions for the incidence of a P-wave or S-wave, both reflected and transmitted P- and Swaves are generated - so called converted waves. They allow, for example, obtaining S-wave information in cases where S-waves are primarily not generated by the seismic source used (see radiation pattern in Fig. 4.6-8). The angle of the reflected or transmitted wave at oblique incidence can be easily calculated using Snell’s law: sin i D (4.6.6) p const. vD
with p being the ray parameter (slowness), iD the angle of the incident, reflected or transmitted ray and vD the (P- or S-wave) velocity of the medium in which the wave is propagating. Snell’s law is the fundamental principle for all ray-geometry observations and describes the reflection and refraction angles between two half-spaces with different velocities. Due to the velocity contrast at the interface the transmitted wave is refracted and changes its direction of propagation. Note that the reflection and refraction angles are independent from bulk density. For example, a P-wave propagating in medium 1 with the velocity vp1 incidents on a planar interface separating medium 1 and 2 under the angle ip1. Due to the above described boundary conditions, four waves arise: one reflected P-wave, one reflected SV-wave, one transmitted P-wave and one transmitted SV-wave (Fig. 4.6-6). The reflected waves propagate with the velocities vp1 and vs1 of medium 1. The refracted waves propagate with the velocities vp2 and vs2 of medium 2. Using Snell´s law (4.6.6) the angles of reflected and refracted rays can be calculated:
sin i p1
sin i p2
sin i s1
sin i s2
v p1
v p2
v s1
v s2
p
const .
(4.6.7)
The premises for reflection and transmission of waves are valid only for planar and adequately extensive interfaces. They are not applicable to the area around faults and in case of truncated reflectors, isolated objects or layers wedging out. If the extension of such structures is equal or less than the wavelength, the energy will neither be reflected nor refracted, but diffracted. Diffractions are explained by Huygens’ principle. When the incident wavefront strikes a point (diffractor), this point is considered to be the source of a secondary spherical wave.
350
4.6 Seismic Methods
Fig. 4.6-6: Reflection, refraction and mode conversion of a P-wave incident at a plane interface
Diffractions play an important role in the mapping of reflectors consisting of several elements of limited extent (fault and fracture zones, horst and graben structures, etc.) or of single objects like cavities, boulders, and man made subsurface objects. Identifying diffracted waves on seismic records may be difficult, since the reflected, transmitted, and diffracted waves interfere with each other.
4.6.3.4 Surface Waves In a homogeneous isotropic full-space only body waves exist, and are transmitted as compressive waves with velocity vp and as shear waves with velocity vs. If the medium is bounded by the Earth’s surface (stress free surface) different types of surface waves (for example Rayleigh, Love, Lamb waves) are also present. Surface waves (boundary waves) are usually characterized by low velocity, low frequency, and large amplitude. They propagate directly from source to receiver, are confined to the surface (interface), and depending on the wavelength their amplitude decreases with depth. At receiver stations close to the source location they tend to obscure the
Environmental Geology, 4 Geophysics
351
events of reflected waves, because of their often very large amplitudes. In conventional seismics they are merely regarded as coherent noise (“ground roll”). For the investigation of near-surface layers, however, they are an important source of information (e.g., HERING et al., 1995; SOCCO & STROBBIA, 2004). Rayleigh waves appear along a free surface and result from interfering Pand SV-waves. Particle motion in a homogeneous medium is retrograde 3 elliptical in the vertical plane and the wavefront is approximately cylindrical. In contrast to spherical body waves (Section 4.6.3.1) their wave amplitudes decay with 1/ r , r being the distance from the source. Within a homogeneous half-space Rayleigh waves are not dispersive. Their velocity vR is slightly less than the velocity of an S-wave (vR | 0.92 vs for Poisson’s ratio V = 0.25 – see Table 4.6-1). However, when a thin layer with lower velocity overlays a homogeneous half-space (e.g., the weathering layer), Rayleigh waves become dispersive. In this case the propagation velocity depends on the frequency or wavelength. Waves with longer wavelengths penetrate deeper into the ground. Waves with shorter wavelengths are sensitive to near-surface material. The dispersive behavior allows determining the thickness and Swave velocity of near-surface layers. XIA et al. (2004) developed methods for display and inversion of Rayleigh waves. The multichannel analysis of surface waves (MASW) includes acquisition, extraction, and inversion of high frequency Rayleigh waves. By inversion of dispersion curves (phase velocities as a function of source-receiver distance), taking into account P-wave velocity, thickness and bulk density, S-wave velocities of near-surface layers can be calculated (XIA et al., 1999). Practical experiences, however, show that severe inhomogeneities in the near-surface layer have an adverse effect on the success of the method. Love-waves (also Q-waves) occur when a free surface and a deeper interface are present and the shear body wave velocity in the near-surface layer is lower than that in the underlying layer. As a surface seismic channel wave their particle motion is parallel to the Earth’s surface and perpendicular to the propagation direction (as for SH-waves). The velocity of Love waves is intermediate between that of the S-wave velocities of the adjacent layers. They always show dispersive behavior, but usually travel faster than Rayleigh waves. The dispersion of Love-waves can be used to determine the thickness of near-surface layers. Because Love-waves are not generated by pure P-wave sources, they can only be used for subsurface investigation when an SH-wave source is applied (Figs. 4.6-1 and 4.6-8).
3 The particle movement is opposite (counterclockwise) to the propagation direction.
352
4.6 Seismic Methods
4.6.3.5 Seismic Resolution In a seismic context, resolution can be defined as the ability to differentiate between two geological features that are close to each other. A distinction must be made between vertical and horizontal (or lateral) resolution, but they are not independent of each other - improving one automatically improves the other. Resolution is determined by the wavelength O of the seismic signal and the target depth z. Short wavelengths (higher frequencies) provide better resolution than longer ones. The wavelength is directly proportional to the wave velocity v in the corresponding material and inversely proportional to the (dominant) frequency f of the seismic signal: O = v / f. Vertical resolution implies how thin a layer can be to allow the separate detection of both its upper and lower boundary. Due to the relatively short near-vertical raypaths reflection seismics offers highest vertical (and horizontal) resolution of seismic methods. The vertical resolution is within the range of O/4 and O/8 (WIDESS, 1973). Up to this value two reflections can be separated and the true layer thickness can be obtained. When the thickness d of a layer is 1000 ms for karst. Since relaxation takes place more rapidly at grain surfaces, the relaxation time is related to the mean pore size and, therefore, grain size of the material. Empirical relationships show that clayey material, including sandy clay, usually has decay times of less than 30 ms, whereas that of sand is of the order of 60 - 300 ms, gravel 300 - 600 ms and pure water about 1000 ms (Fig. 4.7-3). Several authors (SCHIROV et al., 1991; LIEBLICH et al., 1994; YARAMANCI et al., 1999) have shown that SNMR measurements can detect only free water in the pores. Adhesive water on the pore surfaces has very short relaxation times and cannot be detected due to the inevitable dead time (Fig. 4.7-2). The phase M0 of the relaxation signal is related to the phase of the excitation signal in Equation (4.7.3). If the conductivity of the ground is negligible, BA (r) is in phase with the excitation current (M0 = 0). If the ground is highly conductive, a secondary magnetic field, superimposed on the primary field is induced. This modifies the amplitude and phase of the field BA (r). Therefore, M0 is an indicator of the conductivity of the ground. The influence of conductivity on the signal amplitude reduces the investigation depth, as is the case for all electromagnetic methods. A highly conductive layer above the aquifer decreases the amplitude of the SNMR signal, the same layer below the aquifer increases it. The resolution and precision of the method depend on the BA (r) gradient and, therefore, decrease with depth. A strong current io and/or long W is needed to excite the protons at greater depths (as long as W 7.5 mm h-1). Precipitation intensity is of particular significance from the hydrological/hydrogeological point of view. Surface runoff is greater for short heavy rains while groundwater recharge is greater in the case of moderate continuous rain (MATTHES & UBELL, 1983; HÖLTING, 1992). The highest amounts of precipitation usually occur within short rainfall events. The highest daily precipitation observed in the Mediterranean area is 400 - 700 mm d-1 and the highest observed in tropical areas and monsoon and typhoon areas is > 1200 mm d-1 (MATTHES & UBELL, 1983). For the hydrological characterization of an area, the mean annual precipitation is an important parameter. Figures 5.2-2 and 5.2-3 show as examples the distribution of mean annual precipitation in Germany and Thailand for a period of 30 years. Of special interest is the spatial variability of the mean annual precipitation related to geographic units. The amount of precipitation depends especially on the latitude and distance from the moisture source. For instance a monsoon coming from the sea brings more precipitation than a monsoon from inland. Moreover, precipitation varies in a region over the year. Figure 5.2-4 shows the annual distribution of monthly mean precipitation in Germany (Cottbus, State of Brandenburg) and Thailand (Chiang Mai Station).
572
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-2: Mean annual precipitation in Germany for the period 1961 - 1990, www.klimadiagramme.de
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
573
Fig. 5.2-3: Mean annual precipitation in Thailand for the period 1969 - 1998, Department of Meteorology, Thailand (1999)
574
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-4: Monthly average precipitation of the climatological normal period 1961 - 1990 for Cottbus, Germany (black) and Chiang Mai, Thailand (gray), www.klimadiagramme.de/Deutschland/cottbus.html and www.klimadiagramme.de/Asien/chiangmai.html
Fig. 5.2-5: Average annual precipitation recorded at 97 weather stations on the Colorado Plateau from 1900 to 2000. Horizontal lines show the average precipitation for three multidecadal precipitation regimes, HEREFORD et al. (2005).
The long-term variability of precipitation must also be taken into account for the hydrogeological investigation and assessment of an area. Figure 5.2-5 shows the long-term variability of annual precipitation on the Colorado Plateau as an example (HEREFORD et al., 2005). The Colorado Plateau is an arid to semiarid region with mean annual precipitation of 136 to 668 mm a-1 and a median precipitation value of 300 mm a-1. Mean annual precipitation, based on analysis of daily records from 97 long-term weather stations, varied
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
575
considerably during the 20th century (Fig. 5.2-5). Three multi-decadal wet and dry regimes are recognizable in the precipitation time series: 1905-1941, 1942-1977, and 1978-1998. HEREFORD et al., 2005 show that the precipitation variability on the Colorado Plateau is linked spatially and temporally with events in the Pacific Ocean. Measurement of precipitation The amount of precipitation can be measured at selected points by recording and non-recording rain gauges (also spelled rain gages), as well as within an area by radar measurements. A rain gauge is an instrument that collects precipitation and measures the accumulated volume during a certain period of time. The precipitation depth is equal to the accumulated volume divided by the collection area of the gauge. So that measurements within an observation network can be compared, standardized gauges must be used and set up in a standardized way (same height above ground, etc.). The values obtained are point values. Several examples of rain gauges are shown in Fig. 5.2-6. Rain gauges are open containers that catch and accumulate the total precipitation. The U.S. Weather Bureau (now: National Weather Service) nonrecording standard rain gauge, for instance, is a cylinder, 8 inches (20.32 cm) in diameter, corresponding to a collecting area of 324.3 cm2. Rain gauges after Hellmann (used e.g., in Germany, Fig. 5.2-6a) have a collecting area of 200 cm2. Rain gauges should be mounted 1 m above the ground at a site uninfluenced by obstacles (e.g., trees, buildings). Solid and liquid precipitation is collected with the same gauge. The gauges must be read daily at a defined time (e.g., in Germany at 7 a.m., or in the United Kingdom at 9 a.m.).
(a)
(b)
(c)
Fig. 5.2-6: Rain gauges: (a) Hellmann sampler, (b) recording rainfall gauge, (c) totalizator rain gauge for measurements over a long period of time, DRACOS (1980) & WECHMANN (1964)
576
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Larger gauges (collecting area 500 cm2) can be installed in mountainous areas above 1000 m MSL. Snow is melted using calcium chloride (CaCl2) and the snow water is measured. Recording gauges do not need to be visited to record the amount of precipitation collected. They can be equipped with a telemetry unit and are useful at remote, rarely visited or poorly accessible sites. An example is the tipping bucket rain gauge. The rain enters the gauge and is directed into a small bucket that holds a defined amount (e.g., 2 mL) of water. When the bucket is filled, it tips over bringing the second bucket under the funnel. Each tip generates an electric impulse, which is recorded and is a measure of rain intensity in increments of 0.1. Float gauges and weighing gauges measure accumulated precipitation by continuously recording, the change in the height of the float or the accumulated weight of water. Totalizators collect precipitation for a long period of time (monthly, half-yearly or yearly). Precipitation measurements are simple in principle. Nevertheless, difficulties occur in the determination of the amount. Errors are mainly due to effects of wind, ignoring trace amounts, evaporation from the gauge, water adhering to the sides of the gauge, and poor gauge exposure. Wind/turbulence error is usually around 5 %, but can be as large as 40 % in high winds (WILSON & BRANDES, 1979). The precision is 10 to 15 % for rain and up to 40 % for snow (BRUCE & CLARK, 1966). Measurements must be made for at least one year, including both dry and wet seasons. The minimum density of one gauge per 25 km2 is recommended, considering that some heavy rain events (e.g., thunderstorms) may cover only an area of 20 km2 (HISCOCK, 2005). Higher rain gauge densities may be necessary in mountainous areas, where orographic effects cause rainfall variations over short distances. The quality of precipitation data increases with an increase in rain gauge density. Weather radar can detect any type of hydrometeors 1 (e.g., rain, snow, clouds, dew, and frost) in atmosphere, therefore it can be used to determine the amount of precipitation in a specific area. An electromagnetic radar pulse (Chapters 3.3 and 4.5) is used to determine the reflection (echo) of the hydrometeors. In general, the intensity of the echo is a measure of the precipitation intensity. The distance from the radar site to the precipitation area can be measured by the traveltime between transmission of the radar pulse and the arrival of the echo. The spatial coordinates of the precipitation can be determined from the direction and distance measurements. These measurements can be disturbed by interference from trees and buildings (ground clutter), wind drift of particles, drop size distributions, and type of storm. Precipitation measurements by radar can cover an area instead of single point measurements by gauges on the ground. The radar estimates of the 1 Any form of liquid or solid water in the atmosphere.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
577
intensity of precipitation must be calibrated by measurements with traditional rain gauges. Therefore, the joint use of radar and gauges leads to more accurate estimates of precipitation (WANIELISTA et al., 1997; LEGATES, 2004). The U. S. National Weather Service operates numerous WSR-88D (formerly NEXRAD) weather radar systems in the United States (HUNTER, 1996; HARDEGREE et al., 2003; LEGATES, 2004). These radar systems use the reflectivity of S-band microwaves (10 cm wavelength) to provide 1-hourly gridded precipitation estimates with high spatial resolution (4 km × 4 km). HARDEGREE et al. (2003) compared 150 rain gauges in six watersheds with radar precipitation estimates using three years of data. In 5 of 6 cases, the radar underestimated ground precipitation and in all cases, the gauge network detected precipitation for a much longer time than the radar. Similar tests were carried out, for example, in Thailand (CHOKNGAMWONG et al., 2005) with comparable results. HUNTER (1996), GERMANN & JOSS (2001) and LEGATES (2004) describe possible causes of errors in radar precipitation estimates and possibilities to improve them. Hydrologic requirement for precipitation data For hydrologic purposes, single point precipitation measurements as well as areal estimates of precipitation are used. For site characterization, single point precipitation data may be used if they are measured within a reasonably close distance (d 15 km) from the site under investigation (SARA, 1993) and there are no local factors affecting precipitation (e.g., orographic barriers) in the area. In numerous cases (e.g., in hydrological modeling) estimation of areal precipitation (e.g., for a catchment area) is necessary. A number of techniques are available to convert single-point precipitation measurements to areal estimates. An example for a catchment area is given in Figs.5.2-7 to 5.2-10 and Table 5.2-2. The simplest method is to calculate the arithmetic mean of the amounts of precipitation recorded for all the rain gauges in the area. The arithmetic mean is acceptable in relatively flat areas where rainfall does not vary drastically with locality (THOMPSON, 1999). Another method is to use Thiessen polygons (Fig. 5.2-8) to define the zone of influence of each rain gauge in the area. The polygons are constructed by first drawing lines between adjacent gauges. Next, the perpendicular bisectors of these connecting lines are drawn. The perpendicular bisectors are then joined to form the polygon boundaries. The areas of the polygons are determined with a planimeter or by digitizing the map and calculating the polygon areas. While this method assigns weights based on the location of the station in the basin and the relative density of the station network, it cannot account for the effects of elevation on precipitation.
578
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
In isohyetal analysis (Fig. 5.2-9), lines of equal precipitation (isohyets) are drawn based on point measurements with rain gauges. Then the precipitation in each grid area is estimated. The values of all the grid areas are added and the sum is divided by the number of grid areas in the investigation area to estimate the mean precipitation. It is assumed that the effects of elevation on precipitation are taken more into account by isohyetal analysis than in other methods of estimation of areal precipitation, because the isohyets follow considerably the topography (e.g., orographic barriers). The MAPX WSR 88-D estimate (Fig. 5.2-10) uses the 1-hourly gridded precipitation weather radar estimates on a 4 km × 4 km grid. The mean areal precipitation for a catchment area is calculated by averaging the grid point estimates. Table 5.2-2 gives a comparison of results of several mean areal precipitation estimates for an example catchment area.
Fig. 5.2-7: Calculation of the mean areal precipitation for an example catchment area (Station A: 0.55” = 14 mm, Station B: 0.87” = 22 mm, Station C: 2.33” = 59 mm, Station D: 5.40” = 133 mm, Station E: 1.89” = 48 mm), after National Weather Service, Arkansas-Red Basin River Forecast Center (2003)
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
579
Fig. 5.2-8: Calculation of mean areal precipitation for an example catchment area, Thiessen polygons, after National Weather Service, Arkansas-Red Basin River Forecast Center (2003)
580
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-9: Calculation of mean areal precipitation for an example catchment area, isohyetal analysis, after National Weather Service, Arkansas-Red Basin River Forecast Center (2003)
Fig. 5.2-10: Calculation of mean areal precipitation for an example catchment area, MAPX WSR 88-D estimate, after National Weather Service, Arkansas-Red Basin River Forecast Center (2003)
581
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
Table 5.2-2: Results of several mean areal precipitation estimates for an example catchment area, after National Weather Service, Arkansas-Red Basin River Forecast Center (2003) Calculation method arithmetic mean Thiessen polygons isohyetal analysis MAXP WSR 88-D estimate
Result [inches] 2.21 2.03 1.90 2.00
Result [mm] 56 52 48 51
Figure no figure 5.2-8 5.2-9 5.2-10
5.2.2 Evaporation and Evapotranspiration FLORIAN JENN, KLAUS KNÖDEL, MANJA LIESE & HANS-JÜRGEN VOIGT
Evaporation is the conversion of a liquid to a vapor at a temperature below its boiling point. It is the opposite of condensation. The term is also used for the quantity of water that is evaporated. Water evaporates from a variety of surfaces, such as lakes, streams, soils and wet vegetation. Transpiration is the process by which water in plants, absorbed through the roots from the soil, is transferred into the atmosphere from the plant surface, principally from the leaves. Almost all water taken up by the plants is lost by transpiration and only a tiny fraction (about 1 %) is used within the plant in the growth process. The phenomena evaporation and transpiration are often combined to evapotranspiration (ET). Evapotranspiration is the water discharged to the atmosphere by evaporation from the soil and surface-water bodies and by plant transpiration. The term is also used for the volume of water lost through evapotranspiration. The evapotranspiration rate is normally measured in millimeters per time unit. The time unit can be a day, month or year. The amount of water lost by evapotranspiration is expressed in units of water depth. Therefore, it can be converted to volume per unit area (e.g., 1 mm d-1 is equivalent to 10 m3 ha-1 d-1). The rate of evapotranspiration from the Earth's surface is governed by the following factors: x Energy availability: Energy is provided first of all by solar radiation and to a lesser extent by the ambient temperature. The more energy that is available, the greater the rate of evapotranspiration. x Vapor pressure gradient (humidity gradient), i.e. the difference between the water vapor pressure at the surface at which evaporation or transpiration is occurring and that of the surrounding atmosphere. The rate at which water
582
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
vapor enters the atmosphere increases when the vapor pressure of the atmosphere decreases (the air becomes drier). x Wind speed at the Earth's surface: If the wind across the evaporation or transpiration surface replaces the air containing the vapor from evaporation or transpiration with drier air, the rate of evapotranspiration is increased. x Water availability: Evapotranspiration cannot occur if water is not available. x Types and condition of vegetation and land use: When there is little vegetation, water is predominately lost by evaporation from the soil, but once the vegetation is well developed and completely covers the soil, transpiration becomes the main process. When a crop is planted, nearly 100 % of ET comes from evaporation; when the ground is fully covered, more than 90 % of ET comes from transpiration (ALLEN et al., 1998). Plants with deep reaching roots can more constantly transpire water. Woody plants keep their structure over long winters while herbaceous plants largely die back during the winter in seasonal climates, and will contribute almost nothing to evapotranspiration in the spring as they regenerate themselves. Conifer forests tend to have much higher rates of evapotranspiration than deciduous forests. This is because their needles have a much larger surface area. x The soil water content, the ability of the soil to conduct water to the roots, waterlogging and soil water salinity are other factors that affect the evapotranspiration. Hydrologists distinguish between actual evapotranspiration and potential evapotranspiration. The term “actual evapotranspiration” is used for evapotranspiration that actually occurs, while potential evapotranspiration is the evapotranspiration that would occur if a sufficient water source is available. Potential evapotranspiration is called “reference crop evapotranspiration” or “reference evapotranspiration” (ETo) by agricultural experts. The concept of reference evapotranspiration was introduced to study the evaporation caused by climatological effects independently of crop type, crop development, management practices, and soil factors if water is abundantly available. The reference surface is a hypothetical grass reference crop with specific characteristics (ALLEN et al., 1998). Examples of actual evapotranspiration and potential evapotranspiration are given in the Figs. 5.2-11 to 5.2-13. Actual evapotranspiration is usually less than precipitation due to percolation or runoff loss. Exceptions are areas with a high groundwater table, where capillary rise can move water to the ground surface and lakes whose tributary/tributaries from areas with elevated precipitation (e.g., mountains). Due to the ability of soil to hold water and a short supply of energy there is normally a time shift between precipitation and evapotranspiration.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
583
Potential evapotranspiration is usually calculated from climate data. It also depends on the surface type (e.g., ocean, lake, bare soil, crop), soil type, and type of vegetation. Often a value for the potential evapotranspiration is calculated for a climate station on a reference surface, conventionally short grass, and is converted to other areas by multiplying by a surface coefficient. If the potential evapotranspiration is greater than the precipitation, then the soil will dry out unless it is irrigated. On a global scale, most of evapotranspiration of water at the Earth's surface occurs in the subtropical oceans. The mean evapotranspiration value for Germany is 532 mm a-1. Depending on soil and land use, actual evapotranspiration varies from less than 350 mm a-1 in urban areas and the higher parts of the Alps, the Erzgebirge mountains and other upland regions, to values above 700 mm a-1, primarily in the Upper Rhine Plain (Oberrheinebene) and areas with a groundwater table close to land surface (BMU, 2003; WINNEGGE & MAURER, 2002).
Fig. 5.2-11: Annual areal actual evapotranspiration in Australia based on a 30 year climatology (1961 to 1990), Bureau of Meteorology of Australia (2001), www.bom.gov.au/cgi-bin/climate/cgi_bin_scripts/evapotrans/et_map_script.cgi
584
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-12: Annual areal potential evapotranspiration in Australia based on 30 a year climatology (1961 to 1990), Bureau of Meteorology of Australia (2001), www.bom.gov.au/cgi-bin/climate/cgi_bin_scripts/evapotrans/et_map_script.cgi
Evapotranspiration is an important part of the hydrologic cycle (Fig. 5.2-1). A large part of precipitation, for example in Germany 62 % (HÖLTING, 1992) and in Australia 90 % (WANG et al., 2001), is returned to the atmosphere by evapotranspiration. This water does not contribute to the groundwater recharge and is a significant loss of water in the water budget of a catchment area. Therefore, it is necessary to estimate the value of ET as well as possible. It is difficult to exactly determine the ET value due to the complexity of the factors influencing evaporation and transpiration. Consequently numerous approximation methods have been developed.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
585
Fig. 5.2-13: Mean annual evapotranspiration in Thailand in last 30 years in mm d-1, the Meteorological Department of Thailand (1999), www.ldd.go.th/FAO/z_th/thmp131.htm
586
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Water budget equation The open water evaporation Eo from lakes can be estimated from the following mass-balance equation: Eo = I – O ± S,
(5.2.2)
where Eo I = P + IS + IG
is open water evaporation, inflow (precipitation + surface inflow + groundwater inflow), O = OS + OG + DP outflow (surface + subsurface outflow + deep percolation into the subsoil), and S is the change in storage. This method has limited application. All errors in the measurement of the individual components are summed in the final estimate of evaporation. In addition, some of the components like subsurface flow are difficult to measure (WATSON & BURNETT, 1993). Evaporation tank or pan Evaporation can be measured with several different devices. But, the amount of evaporation measured with different devices is usually not the same. The U.S. National Weather Service evaporation pan called a Class A pan is accepted as an international standard. A Class A evaporation pan is cylindrical with a diameter of 47.5 inches (1206.5 mm) and a depth of 10 inches (254 mm) made of stainless steel. The pan is placed on a carefully leveled, wooden base on the ground in a grassy location and is often enclosed by a chain link fence to prevent animals drinking from it. The Sunken Colorado pan (square pan) and the British standard tank, both of which are 1.83 m × 1.83 m and a water depth 0.61 m, are also commonly used. Tank evaporimeters tend to give more accurate measurements of true evaporation, because tanks are less affected by heating of the walls than pans (HISCOCK, 2005). Pan evaporation is a measurement that combines the effects of temperature, humidity, solar radiation, and wind. Evaporation is measured daily as the depth of water decreases in the pan. The evaporation rate can be measured by manual readings or with an evaporation gauge with analog output. Since the amount of evaporation is a function of temperature, humidity, wind, and other conditions, the maximum and minimum temperature of the water and the amount of air passage are normally recorded along with the evaporation. Water level sensors, water temperature sensors, weather sensors, and data-logging capabilities can be part of an evaporation monitoring system. The measurement day begins each morning at 7 o'clock with the pan filled to exactly two inches (50.8 mm) from the top of the pan. After 24 hours the remaining quantity of water is measured and the amount of
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
587
pan evaporation per time unit Epan (mm d-1) is obtained from the difference between the two measured water depths. If precipitation occurs in the 24-hour period, it is measured simultaneously and taken into account in calculating the evaporation. Sometimes precipitation is greater than evaporation, and measured amounts of water must be removed from the pan. The potential evapotranspiration ETo is calculated by multiplying Epan by the pan coefficient Kpan: ETo = Kpan u Epan
(5.2.3)
For the Class A evaporation pan, Kpan varies depending on season and location from 0.35 to 0.85. The average Kpan is 0.70. Determination of actual evapotranspiration with lysimeter Lysimeters are tanks filled with either disturbed or undisturbed soil buried with tops flush with the ground level. The soil in them should be planted with same type of vegetation as the surrounding area. Two types of lysimeters are available, weighing (Fig. 5.1-14) and nonweighing. In precision weighing lysimeters, the water loss is directly measured by the change of mass. Evapotranspiration can be obtained with an accuracy of a few hundredths of a millimeter, and small time periods such as an hour can be considered (ALLEN et al., 1998). In nonweighing lysimeters the ET for a given time period is determined from the difference between the total water input and the drainage water, collected at the bottom of the lysimeters. Nonweighing lysimeters should be used to determine the actual evapotranspiration only if the change in soil moisture is negligible.
Fig. 5.1-14: Diagram of a weighing lysimeter, DYCK & PESCHKE (1989)
588
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Precipitation must be measured simultaneously. The evapotranspiration can be estimated with the following water-balance Equation: ET = P + IR – DP ± ǻS, where ET P IR DP ǻS
(5.2.4)
is the evapotranspiration per time unit, precipitation, irrigation water added to the lysimeter, deep percolation (excess water drained from the bottom of the lysimeter), and soil moisture storage.
As lysimeters are difficult and expensive to construct, operate and maintain, their use is limited to research purposes (ALLEN et al., 1998). A nonweighing lysimeter in combination with a neutron probe can be used to determine evapotranspiration. A nonweighing lysimeter is used to determine deep percolation and the neutron probe measures the soil moisture. When all other parameters (rainfall, irrigation, surface and subsurface inflow and outflow) are measured or reduced to zero, the ET can be determined, using the water-balance Equation (5.2.4). Thornthwaite equation In the Thornthwaite equation the mean monthly temperature is used to estimate potential evapotranspiration. ET
§ 10 t · 1.6 ¨ ¸ © TE ¹
a
(5.2.5)
a = 0.49239 + 0.01792 TE TE
1.514 º ¦ ª t i / 5 »,
12
¬ i 1«
where ET t TE
¼
(5.2.6) (5.2.7)
is evapotranspiration (cm per month), mean monthly temperature (°C), and Thornwaite´s temperature efficiency index.
This equation usually has to be adjusted for the month and latitude. Blaney-Criddle equation This method for calculating potential evapotranspiration is based on the monthly mean temperatures and the monthly mean percentage of daylight hours per day. It can be used if no locally measured data on pan evaporation are available. The Blaney-Criddle equation (BROUWER & HEIBLOEM, 1986) is
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
ET = p (0.46 Tmean + 8), where ET Tmean p
589 (5.2.8)
-1
evapotranspiration (mm d ) as the average for a period of one month, mean daily temperature (°C), and mean daily percentage of annual daytime hours.
The mean annual percentage of daylight hours for the latitudes 0° to 60° N and S and the months of the year can be taken from Table 12 of BROUWER & HEIBLOEM (1986). The method only provides a rough, order-of-magnitude estimate. Especially in windy, dry, sunny areas, ET is underestimated by up to about 60 %, while in calm, humid, clouded areas, the ET is overestimated by up to about 40 % (BROUWER & HEIBLOEM, 1986). Turc equation The Turc equation give approximate values of actual evapotranspiration averaged over a year based on mean annual precipitation and mean annual air temperature: ET
P 0.9 P / L
where ET P L T
2 1/ 2
,
(5.2.9)
evapotranspiration in mm per year, mean annual precipitation in mm, = 300 + 25 T + 0.05 T 3, and mean annual air temperature in °C
This equation can be used for almost all climates. FAO Penman-Monteith method The Penman-Monteith method is recommended by the Food and Agriculture Organization of the United Nations (FAO) as the sole standard method with the reason that the method is able to predict the reference crop evapotranspiration ETo correctly for a wide range of locations and climates. The reference crop is defined “as a hypothetical crop with an assumed height of 0.12 m having a surface resistance 2 of 70 s m-1 and an albedo of 0.23, closely resembling the evaporation of an extension surface of green grass of uniform height, actively growing and adequately watered…” (ALLEN et al., 1998). For the calculation of ETo using Equation (5.2.9) after ALLEN et al., (1998), standard climatological records of solar radiation (sunshine), air temperature, humidity and wind speed are required: 2 aerodynamic resistance
590
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
900 u 2 es e a T 273 , ' J 1 0.34 u 2
0.408 ' R n G J ETo
where ETo Rn G T u2 es ea es - ea ǻ
J
(5.2.10)
reference evapotranspiration [mm d-1], net radiation at the crop surface [MJ m-2 d-1], soil heat flux density [MJ m-2 d-1], mean daily air temperature 2 m above the ground [°C], wind speed 2 m above the ground [m s-1], saturation vapor pressure [kPa], actual vapor pressure [kPa], saturation vapor pressure deficit [kPa], slope of the vapor pressure curve [kPa °C-1], psychrometric constant [kPa °C-1].
For more details, examples, parameter tables, and how this method can be applied when there is little data, see ALLEN et al., (1998). Other methods The use of Landsat images define crop distribution and thermal NOAAAVHRR images to determine spatial variation in temperature has been proposed by HURTADO et al. (1997) for estimating actual evapotranspiration. Actual daily evapotranspiration can be estimated with this with an error of no more than 0.9 mm d-1.
5.2.3 Runoff FLORIAN JENN, KLAUS KNÖDEL, MANJA LIESE & HANS-JÜRGEN VOIGT
After precipitation and evapotranspiration the third component of the water balance of a catchment area is runoff. Runoff is that part of precipitation that flows toward streams on the surface of the ground or within the ground. Runoff may be classified according to how quick it appears after rainfall or snow melts (quick flow or baseflow) or according to location (surface runoff, interflow, or groundwater runoff). Total runoff (R) is composed of overland flow (RSu), interflow (RI), and groundwater runoff baseflow (RB): R = RSu + RI + RB
(5.2.11)
Overland flow is that portion of precipitation, snow melt, or irrigation water in excess of what can infiltrate the soil surface or be stored in small surface depressions. It is a major transporter of non-point source pollutants in rivers,
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
591
streams, and lakes. Rain that falls on the stream, usually less than 1 % of runoff, is often included in this parameter. This channel precipitation is only significant where there is a large area of wetlands, lakes or during floods, but it always occurs. That portion of rainwater or snowmelt which cannot infiltrate the soil moves as overland flow over the land surface toward stream channels. This overland flow is expressed as discharge per unit width of slope. If the intensity of the rain is less than the infiltration rate, all precipitation water will infiltrate and no overland flow occurs. Three types of overland flow are distinguished: x Hortonian overland flow, x saturation overland flow, and x overland flow due to surface crusting. Hortonian overland flow occurs if the rain intensity exceeds the infiltration rate and the surplus water becomes runoff. Initially, depth and velocity are very small and sheet flow prevails. When the velocity increases, rill flow (flow in small rivulets) occurs. When velocity and depth of overland flow increase, soil erosion results and contaminant transport increases. It is more frequent in arid areas, where the vegetation is sparse and the infiltration rate is low. Saturation overland flow occurs if the soil is water-saturated. This may happen after a long precipitation event or if the groundwater table is shallow, for example adjacent to a stream. This is more often the case in humid areas. Overland flow due to surface crusting occurs when the surface becomes dry and crusted. This is mostly the case in arid areas, where the soil dries out (among other things because there is no vegetation cover protecting it) during the dry season and becomes encrusted. Interflow is that portion of rainfall or snowmelt that infiltrates into the soil and moves laterally through the unsaturated zone during or immediately after a precipitation event and discharges into a stream or other body of water. This usually takes longer to reach a stream than surface runoff. Rapid interflow occurs as macropore flow, for example, through root canals and tunnels made by animals. Overland flow dominates in arid areas with poor vegetation and relatively thin soil layer. In humid areas with vegetation and a thick permeable soil layer, interflow predominates. Overland flow and interflow represent the surface runoff or quick flow or direct runoff of a stream catchment area. Groundwater runoff (baseflow) is that part of runoff which has passed into the ground, has become groundwater, and has been discharged into a stream as spring or seepage water. Baseflow is an important source of streamflow between rainstorms. It is estimated by employing a hydrograph separation method in which long-term streamflow records are reconstructed so that most of the stormwater flow is subtracted from those records to theoretically arrive at the groundwater contribution to the stream. This is also referred to as dry-
592
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
weather flow. In catchment areas with permeable sediments, e.g., sand, limestone, or sandstone, the baseflow component is a large fraction of the total runoff. The opposite is true in catchment areas with clay-rich sediments or outcropping hard rock.
Fig. 5.2-15: Mean annual runoff depth for the period 1961 - 1990 in Germany with grid cells representing 1 km2 each, BMU (2003)
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
593
Mean annual runoff in Germany is about 38 % of the mean annual precipitation (HÖLTING, 1992). The mean annual runoff in Germany for the period 1961 - 1990 is shown in Fig. 5.2-15 with grid cells representing 1 km2 each. The values are less than 100 mm a-1 in northeastern Germany and more than 2000 mm a-1 in the higher regions of the Alps (WINNEGGE & MAURER, 2002). Table 5.2-3 gives typical values for continental runoff. Surface runoff sends 7 % of the land based precipitation back to the ocean (PIDWIRNY, 2005). Table 5.2-3: Continental runoff values, LVOVITCH (1972), quoted from PIDWIRNY (2005) Continent Europe Asia Africa North and Central America South America Australia, New Zealand and New Guinea Antarctica and Greenland
Runoff per year [mm a-1] 300 286 139 265 445 218 164
The following factors affect runoff: x type of precipitation (rain, snow, sleet, etc.), x intensity, amount, and duration of a rainfall event, x distribution of rainfall within the watershed, x direction of storm movement, x antecedent precipitation and resulting soil moisture, x other climatic conditions that affect evapotranspiration, such as temperature, humidity, solar radiation, and wind, x land use, x vegetation, x soil type, x soil sealing, x drainage area, x shape and orientation of the catchment area, x topography, elevation and slope, x drainage network patterns, x ponds, lakes, reservoirs, sinks, etc. in the basin, which prevent or alter runoff from continuing downstream.
594
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Determination of streamflow Streamflow (Q) is the amount of water leaving the catchment area, i.e. surface runoff, interflow and baseflow. But these three kinds of flow are difficult to separate. The flow is now analyzed as a function of time, not the flow path (see Section “Runoff separation by hydrograph analysis”). Therefore, streamflow is determined as the volume of water passing a given point per time unit in a stream channel, usually expressed in cubic meters per second (m3 s-1), smaller flows in liters per second (L s-1). Streamflow is a function of depth, width, and velocity of the water in a channel. The streamflow is not constant. During a precipitation event in temperate areas it increases and reaches a maximum after the precipitation event (Fig. 5.2-20). Then it devreases until only baseflow occurs. In arid regions no runoff exists during the season of no precipitation. Ephemeral streams are typical in these regions because there is no baseflow. The variability of runoff during the year in temperate regions is low in comparison to arid regions. Moreover, runoff varies from year to year. During a dry period the soil dries out, and soil moisture decreases. The soil moisture is replenished by the precipitation event that follows. In spite of high precipitation, runoff can be low (because runoff depends on soil moisture, infiltration capacity and precipitation intensity). Several methods can be used to determine streamflow. Some frequently used methods are described here. The simplest method for estimating small flows is by direct measurement of the time to fill a container of known volume. If the flow of water can be diverted into a pipe so that it is discharged under pressure and the pipe is discharges vertically upwards, the height to which the jet of water rises above the end of the pipe can be measured and the rate of flow can be calculated using the appropriate formula (HUDSON, 1993). In another method Q is calculated from the average flow velocity (v) and the cross-section area (A) of the channel: Q = A v.
(5.2.12)
A simple possibility for estimating the velocity is to measure the time taken for a floating object or a strongly colored dye to travel a measured distance downstream. The velocity is not the same at all places in the stream. It is slower at the sides and bottom, and faster on the surface. Taking 0.8 of the surface velocity as measured by the float gives an approximate value for the average velocity (HUDSON, 1993). In turbulent streams a tracer can be added to the stream at a constant measured rate and samples are taken at points downstream. The dilution is a function of the rate of flow which can be calculated. This method is particularly useful when the cross-section of the stream is difficult to determine (ice-covered streams, sand-channel streams or boulder-filled mountain streams).
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
595
More accurate determination of velocity can be obtained with a current meter. A current meter is a device in hydrology for measuring the speed and direction or speed alone of flowing water. Current meters can be mechanical, electric, electromagnetic, acoustic, or combination thereof. Commonly used current meters are vertical-axis Price or Woltmann current meters (Fig. 5.2-16), and the Pygmy meter for shallow channels. For a selected location in a stream, velocity is measured with a current meter at 0.2 and 0.8 of the depth D in several vertical section of the stream (at 0.6 in shallow streams). The streamflow is calculated from the mean velocity of each section and the area of that section using Equation (5.2.12) as shown in Fig. 5.2-17. Streamflow can be monitored by continuously measuring the flow rate with a current meter and the height of the water surface (called stage) above an arbitrarily established datum plane with a water-level recorder. The results are shown on a stage-discharge curve, which is a graph of flow volume versus depth. The relationship between stage and flow rate is affected by the irregularities and changes in the channel resulting from erosion and deposition. For this reason gauging the flow in a natural stream can never be precise. More reliable estimates can be obtained when the flow is passed through a standard structure, such as a weir or flume, constructed for this purpose. Master discharge tables or graphs can be be used to directly obtain the flow in a standard weir or flume. Nonstandard structures must be calibrated after installation.
Fig. 5.2-16: Woltmann current meter, LAWA (1991)
596
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
1 Section 1 2 3 4 5 6 7
2 3 4 Flow velocity [m s-1] 0.2 D 0.8 D Mean 0.5 0.8 0.6 0.7 0.9 0.6 0.75 1.1 0.7 0.9 1.0 0.6 0.8 0.9 0.6 0.75 0.55
5 Depth [m] 1.3 1.7 2.0 2.2 1.8 1.4 0.7
6 Width [m] 2.0 1.0 1.0 1.0 1.0 1.0 2.0
7 8 Flow Area [m3 s-1] [m2] 2.6 1.30 1.7 1.19 2.0 1.50 2.2 1.98 1.8 1.44 1.4 1.05 1.4 0.77 Total 9.23
D is the depth of the stream at the mid-point of each section. Fig. 5.2-17: Estimating streamflow from current meter measurements, HUDSON (1993)
Fig. 5.2-18: Examples for weirs, (a) rectangular-notch weir, (b) 90° V-notch weir
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
597
A weir is a low overflow type dam built across a stream to raise its level or to direct or regulate flow. It can be used to measure the discharge by having the water flow over a specifically designed spillway (FETTER, 2001). Flumes are channel-type structures. The V-notch and the rectangular-notch weirs (Fig. 5.2-18) are the two most common types of weirs. They are portable and simple to install in either temporary or permanent positions. The rectangular notch is more suitable for large flows because the width can be easily adapted to the expected flow. There must be an approach channel on the upstream side to remove any turbulence and ensure that the water approaches the notch smoothly. For accurate measurements the width of the approach channel must be 8times the width of the notch, and it must extend upstream for 15times the depth of flow over the notch and the notch must have a sharp edge on the upstream side. To determine the depth of flow through the notch, a staff gauge is placed in the approach channel well back from the notch so that it is not affected by the drawdown curve as the water approaches the notch. The zero of the scale on the staff is set level with the lowest point of the notch (HUDSON, 1993). The flow rate over a weir can be estimated using the following empirical equations (FETTER, 2001) or from the Table 5.2-4: rectangular-notch weir: Q = 1.84 (L – 0.2 H) H2/3, 90° V-notch weir: where Q L H
Q = 1.379 H5/2, 3 -1
is flow [m s ], length of weir crest [m], and height of the backwater above weir crest [m].
The equations are empirical and not subject to dimensional analysis.
(5.2.13) (5.2.14)
598
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Table 5.2-4: Flow rates over a rectangular-notch weir with end contractions and over a 90° V-notch weir, USDI (1975) quoted after HUDSON (1993) Rectangular-notch weir head [mm] 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 370 380
-1
flow [L s per meter of crest length] 9.5 14.6 20.4 26.7 33.6 40.9 48.9 57.0 65.6 74.7 84.0 93.7 103.8 114.0 124.5 136.0 146.0 158.5 169.5 181.5 193.5 205.5 218.5 231.0 244.0 257.5 271.0 284.0 298.0 311.5 326.0 340.0 354.0 368.5 383.5 398.0
90° V-notch weir head [mm]
flow [L s-1]
40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 350
0.441 0.731 1.21 1.79 2.49 3.34 4.36 5.54 6.91 8.41 10.2 12.0 14.1 16.4 18.9 21.7 24.7 27.9 31.3 35.1 38.9 43.1 47.6 52.3 57.3 62.5 68.0 100.0
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
599
Open channel flow The semi-empirical Manning equation, first presented in 1889 by the Irish engineer Robert Manning, is the most commonly used equation to analyze water flow in channels and culverts where the water is open to the atmosphere, i.e. not flowing under pressure. The flow Q can be determined from the average velocity v of the water flowing in the channel and the cross-section area of the channel A according to Equation (5.2.12). The velocity of water flowing in an open channel is affected by the following factors: x the channel gradient, x roughness of the channel, causing frictional resistance to the moving water in the stream bed, and x shape of the channel, which influences the frictional resistance. The Manning equation takes into account these parameters to determine the velocity of water flowing in an open channel: v
1 2 / 3 1/ 2 M S n
(5.2.15)
The parameter M takes the shape of the channel into account and is called the hydraulic radius of the channel. It is defined as the cross-sectional area divided by the wetted perimeter, which is the width of the bed and sides of the channel which are in contact with the water. Channels with the same cross-section area can have different values for the hydraulic radius M (Fig. 5.2-19). The hydraulic radius thus has units of length. M
A , P
where Q v A M S P
n
(5.2.16) is flow [m3 s-1], average velocity [m s-1], cross-section area of the channel [m2], hydraulic radius [m], slope of the channel [m m-1], wetted perimeter, which is often approximated by the top width of the channel or by top width of the channel + 2 times the depth of the channel [m], and Manning’s coefficient of roughness.
600
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-19: Examples of channels with the same cross-sectional area that have different values for the hydraulic radius M
Table 5.2-5 gives examples of Manning’s roughness coefficients n. Table 5.2-5: Examples of Manning’s roughness coefficients n for streams/channels, FETTER (2001) mountain streams with rocky beds winding natural streams with weeds natural streams with little vegetation straight, unlined Earth canals smoothed concrete
0.04 - 0.05 0.035 0.025 0.020 0.012
Runoff separation by hydrograph analysis The runoff in a catchment area consists of direct runoff and baseflow as shown in Fig. 5.2-20a. A hydrograph from a rainfall event (shown schematically in Fig. 5.2-20b) can be used to estimate the amounts of direct runoff and baseflow. A hydrograph is graphical representation showing for a given point on a stream the discharge, stage (depth), velocity, or other property of water as a function of time. Infiltration and percolation during and after a rainstorm recharges the groundwater, increasing the rate of baseflow, as shown in Figs. 5.2-20b and 5.2-21. For a catchment area of 100 km2, maximum baseflow is reached 3 to 4 days after a precipitation event (RICHTER & LILLICH, 1975). After infiltration, the baseflow declines, as shown in the baseflow depletion curve in Fig. 5.2-20b. Baseflow RB can be estimated from hydrographs as shown in Figs. 5.2-20b and 5.2-21. RB can be calculated as a mean value Qmean from single measurements Qi made during dry periods and by integration of the area below the dry-weather flow line. The area above the dry-weather flow line is assigned to direct runoff. Other hydrograph separation methods are available (RICHTER & LILLICH, 1975; HÖLTING, 1992; HISCOCK, 2005).
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
601
Fig. 5.2-20: Runoff in a catchment area: (a) runoff components, (b) schematic hydrograph from a rainfall event and its interpretation as described in the text (HISCOCK, 2005)
Fig. 5.2-21: Hydrograph of discharge Q as a function of time t of a stream and estimation of baseflow RB as a mean value Qmean from single measurements Qi made during dry periods and by integration of the area below the dry-weather flow line (gray area), RICHTER & LILLICH (1975)
602
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Maximum rate of runoff The probable maximum rate of runoff in a catchment area can be estimated using the following equation (HUDSON, 1993): Q
c I A, 360
where Q I A c
(5.2.17) is rate of runoff [m3 s-1], intensity of rain [mm h-1], catchment area in hectars, and runoff coefficient [dimensionless].
The rain intensity I can be taken from local rainfall records. It is also necessary to take into account the longest time taken by surface runoff to reach the outlet from any point in the catchment area. The coefficient c is a measure of the proportion of the rain which becomes runoff. Values for different surface soils are given in Table 5.2-6. Table 5.2-6: Values of the runoff coefficient c, SCHWAB et al., (1981), quoted after HUDSON (1993) Topography and vegetation woodland flat 0 - 5 % slope rolling 5 - 10 % slope hilly 10 - 30 % slope pasture flat rolling hilly cultivated flat rolling hilly urban areas flat rolling *of area impervious
Soil texture open sandy loam
clay and silt loam
tight clay
0.10 0.25 0.30
0.30 0.35 0.50
0.40 0.50 0.60
0.10 0.16 0.22
0.30 0.36 0.42
0.40 0.55 0.60
0.30 0.40 0.52 30 % * 0.40 0.50
0.50 0.60 0.72 50 % * 0.55 0.65
0.60 0.70 0.82 70 % * 0.65 0.80
603
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
Probable runoff quantity of a catchment area The probable volume of runoff or yield of a catchment area can be calculated for an individual storm using the formula of the U. S. Soil Conservation Service (HUDSON, 1993): Q
P
0.2 S
P 0.8 S
where Q P S
2
,
(5.2.18)
is runoff [mm], storm rainfall [mm], and amount of rainfall which can soak into soil during the storm [mm].
Table 5.2-7 gives values of S [mm] for the water yield Equation (5.2.18). Intermediate values may be used. Table 5.2-7: Values of S [mm] for the water yield Equation (5.2.18), USDA-SCS (1964), quoted after HUDSON (1993) Soil type good permeability, for example, deep sands medium permeability, for example, sandy clay loams and clay loams low permeability, for example, clays
Number of days since last storm which caused runoff more than 5 150
2 to 5 75
less than 2 50
100
50
25
50
25
25
5.2.4 Infiltration FLORIAN JENN, KLAUS KNÖDEL, MANJA LIESE & HANS-JÜRGEN VOIGT
Infiltration is the process by which water at the ground surface passes into the unsaturated, or vadose zone, as well as the volume of that water. More than two thirds of the world’s precipitation on land infiltrates into the ground (SERRANO, 1997). Infiltration provides the water for plant growth and almost all water in the groundwater reservoirs. Infiltration is also important for the spreading of contaminants. Therefore, an understanding of infiltration and water movement through the unsaturated zone is very important in hydrology and hydrogeology. Infiltration is governed by two forces: gravity and capillary action. Temporal and spatial factors greatly affect how water moves through the subsurface. Infiltration depends on the characteristics of the ground surface, such as slope, vegetation cover, land use, soil properties such as soil
604
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
texture, porosity, soil water content, soil chemistry, as well as meteorological factors, such as antecedent rain events, precipitation intensity, depth of ponding, etc. Secondary pores (e.g., shrinkage cracks, root courses, earthworm courses) are also of a high importance. During a rain storm event, the precipitation divides into infiltration, depression storage (ponding), and overland flow (Section 5.2.3) as shown in Fig. 5.2-22. Infiltration begins when the precipitation reaches the ground. Overland flow does not begin until the available depression storage is exhausted. Overland flow can continue past the termination of precipitation. Infiltration will continue as long as there is any water in the depression storage, usually past the period of overland flow.
Fig. 5.2-22: Incremental precipitation rate and its dissociation into amounts of infiltration, depression storage (ponding), and overland flow, http://jan.ucc.nau.edu/~doetqp-p/courses/env302/lec4/LEC4.html
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
605
Fig. 5.2-23: Schematic of infiltration rate dependency on rain intensity i and profiles of water content for (a) low, (b) moderate, and (c) intense precipitation, Ks is the saturated hydraulic conductivity, R recharge, Ti initial water content, T0 water content behind the wetting front, Ts saturated water content, STEPHENS (1996)
Four infiltration scenarios are distinguished: x Low precipitation: The rainfall intensity is less than the magnitude of saturated hydraulic conductivity 3 of the soil. All rainfall infiltrates, no ponding and no overland flow occur (Fig. 5.2-23a). The infiltration rate is equal to the water input rate and is less than the infiltration capacity. x Moderate precipitation: Rain falls on dry soil with a constant rate. The rainfall rate exceeds the magnitude of saturated hydraulic conductivity of the soil. The soil initially takes up the rainwater as rapidly it falls. All rainwater infiltrates until saturation of the uppermost soil layer is reached. Then ponding and overland flow occur and the infiltration rate decreases to the value of infiltration capacity (Fig. 5.2-23b). x Intense precipitation (saturation from above): When rainfall intensity is high relative to the saturated hydraulic conductivity of the soil, ponding and runoff occur instantly and the infiltration declines exponentially and asymptotically to a near constant value corresponding to the magnitude of saturated hydraulic conductivity of the soil (Fig. 5.2-23c). The infiltration rate is equal to the infiltration capacity. 3 Saturated hydraulic conductivity is a quantitative measure of a saturated soil’s ability to transmit water when subjected to a hydraulic gradient.
606
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
x Saturation from below: The water table has risen to or above the surface, ponding, no infiltration occurs. Figure 5.2.23 also schematically shows the water content profiles and the downward propagation of the wetting front of infiltrating water during a rainfall event. The wetting front is diffuse in clay (due to high capillarity) and sharp in sand and, in addition, lateral inhomogeneities of soil physical properties cause an uneven front. The actual infiltration rate is difficult to measure because of the variety of factors it depends on. Analog to the approach in modeling evapotranspiration, in which the two concepts actual “evapotranspiration” and “potential evapotranspiration” are used, the models of actual infiltration rate and infiltration capacity are used. The actual infiltration rate is the quantity of water on a soil surface that enters the soil in a specified time interval. The unit is depth of water per unit time, for example mm h-1 or cm s-1. Infiltration capacity is the maximum rate at which infiltration can occur under given conditions. It is a function of soil moisture content. The unit is depth of water per unit time, for example mm h-1 or cm s-1. The actual infiltration rate is less than the infiltration capacity if no ponding/flooding occurs and is the minimum between infiltration capacity and precipitation rate. The infiltration rate is usually measured for ponded water. The infiltration rate depends on the ponding depth, primarily at early time of the precipitation event. It is adequate to use the term “ponded infiltration rate”. The ponded infiltration rate curve is also affected by the antecedent water content. The dryer the soil, the higher is the initial infiltration rate. Various researchers have modeled the infiltration curve as shown in Fig. 5.23c. One model is the Horton equation for infiltration capacity (HORTON, 1940): fp
f c f 0 f c e kt ,
where fp t f0 fc k
(5.2.19)
infiltration capacity [m s-1 or mm h-1], time [s or min], initial (i.e. maximum) infiltration capacity [m s-1 or mm h-1], equilibrium infiltration capacity (asymptotic infiltration capacity) [m s-1 or mm h-1], a constant, representing the rate of decrease of infiltration capacity [s-1 or min-1].
The parameters f0, fc, and k can be determined from infiltration measurements. For rough estimates of infiltration capacity the parameters from Table 5.2-8 can be used.
607
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations Table 5.2-8: Parameter estimates for Horton’s infiltration model, ASCE (1996) Soil and cover standard agriculture (bare) standard agriculture (turfed) peat fine sandy clay (bare) fine sandy clay (turfed)
f0 [mm h-1]
fc [mm h-1]
k [min-1]
280 900 325 210 670
6 - 220 20 - 290 2 - 29 2 - 25 10 - 30
1.6 0.8 1.8 2.0 1.4
The most widely referenced model is Philip’s transient infiltration equation for very shallow ponded water (PHILIP, 1957; STEPHENS, 1996): i
1 1/ 2 st Ko , 2
where i t s Ko
(5.2.20)
infiltration rate [m s-1], time since infiltration began [s], sorptivity of the soil [cm s-1/2], hydraulic conductivity of the uppermost soil layer [m s-1],
Sorptivity is a soil hydraulic parameter that includes several factors and represents the capacity of soil to take up infiltrating water early during a precipitation event, such as the initial water content of the soil. Philip’s infiltration equation can be used to predict the transient infiltration rate. Soil can be classified by its texture (see Section 5.2.7.3). The texture is determined by the proportions of clay, silt and sand after particles larger than sand have been removed. Clay is defined as soil particles with a diameter less than 0.002 mm. Silt is soil particles between 0.002 and 0.05 mm in diameter and sand is particles between 0.05 and 2 mm in diameter. Particles greater than 2 mm in diameter are classified as gravel. If a significant portion of the soil is greater than 2 mm, the term “gravelly” or “stony” is added to the soil description. The grain-size distribution must be determined for the classification of a soil. Several soil hydraulic properties depend on soil texture. The size of the pores through which water can flow is approximately equal to the grain size. Therefore, the pore-size distribution can be derived from the grain-size distribution. Typical curves of accumulated infiltration for different soil textures and land use are shown in Fig. 5.2-24. Water flows in both the unsaturated, or vadose zone and in the saturated zone from areas of high potential energy to areas with low potential energy. The total soil-water potential consists of two parts: soil-water potential and elevation potential. For the soil-water potential, capillary and adsorptive forces are the most important. When the soil-water potential is expressed in energy units per unit weight of soil, it is called pressure head \ (STEPHENS, 1996). It is also called suction or tension head and is measured with a tensiometer as water column.
608
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-24: Typical infiltration curves for different soil textures and land use, MCLAREN & CAMERON (1990) quoted after MAF (2002)
Pressure head and hydraulic conductivity govern the rate of unsaturated flow in a porous media (soil, sediment, bedrock) according to Darcy’s law. Both pressure head and hydraulic conductivity are related to the water content. The relationship between pressure head (soil water potential) and water content is shown in Fig. 5.2-25 for material with different textures. The water content of soil at a given pressure head depends on its wetting history. This effect is called hysteresis and is depicted in Fig. 5.2-26. When the soil wets from an initially dry condition, the soil behaves according to the main wetting curve. The behavior of drying from complete saturation is depicted as main drying curve. The intermediate paths on the soil-water characteristic curve, bounded by the main wetting curve and main drying curve, are called scanning curves. The factors contributing to hysteresis are capillary effects and the effect of entrapped air in the soil pores.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
609
Fig. 5.2-25: Effect of texture on the soil-water characteristic curve; \ is the pressure head, T is water content (STEPHENS, 1996)
Fig. 5.2-26: Water content versus pressure head (SERRANO, 1997)
610
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-27: Pressure head as a function of depth for the unsaturated and saturated zones, SERRANO (1997)
An idealized representation of the pressure head as a function of depth is shown for a typical soil profile in Fig. 5.2-27. The pressure head at an unconfined water table is zero, and in the zone of saturation the pressure head is positive. In the unsaturated zone the pressure head is negative. The effect of layering on water-content and pressure-head profiles under hydrostatic equilibrium is shown in Fig. 5.2-28. Infiltration ceases when no water from precipitation, lakes, ponds, and streams is available to move through the soil surface into the soil. But movement of water within the soil does not stop when infiltration ceases. After a period of infiltration, water moves in the vadose zone as drainage and redistribution. Drainage in the uppermost vadose zone occurs when the water content decreases over time (Fig. 5.2-29a). Water behind the wetting front can move upward in response of evapotranspiration or it can move downward due to gravity and other potential gradients resulting from capillary, imbibitional and adsorptive forces (Fig. 5.2-29b). This process is called redistribution. Water percolating downward through the vadose zone can reach the water table and contribute to groundwater recharge. According to isotope measurements in various regions, the mean infiltration velocity for Northern Germany is about 1 m a-1 (HÖLTING, 1992).
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
611
Fig. 5.2-28: Effect of layering on water-content and pressure head profiles under hydrostatic equilibrium, \ pressure head, T water content, STEPHENS (1996)
Fig. 5.2-29: Drainage (a) and redistribution (b) following a period of infiltration, t1, t2 and t3 time, where Ti is the initial water content, STEPHENS (1996)
612
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Infiltration is usually quantified by water budget analysis, laboratory and field measurements, and/or calculation based on soil hydraulic properties. Various analytical models are used in soil science to estimate the annual infiltration rate based on the water balance equation. The infiltration rate is calculated from the difference between the annual precipitation and the sum of direct runoff and evapotranspiration. Examples of such models in Germany are ABIMO (GLUGLA & FÜRTIG, 1997), GROWA (WENDLAND et al., 2001), and the RENGER-WESSOLEK (RENGER & WESSOLEK, 1996). Lysimeters can also be used to study infiltration. For their operation and limitations, see Section 5.2.2 and BREDENKAMP et al. (1995). Other methods are described in Section 5.2.7.
5.2.5 Groundwater Recharge FLORIAN JENN, KLAUS KNÖDEL, MANJA LIESE & HANS-JÜRGEN VOIGT
Groundwater recharge is the process of water infiltrating through the vadose zone and reaching the groundwater surface, thus adding water to the aquifer. The amount of this water is also called “groundwater recharge”. Groundwater recharge occurs naturally as the net gain from precipitation or runoff. Injection of water into an aquifer through wells is one form of artificial recharge. Another method is to direct runoff into a collecting pond with a permeable bottom; this is often done in areas with hardpan. The groundwater recharge rate is reported in mm a-1 or L s-1 km-2 and the volume of groundwater recharge is calculated from the recharge rate multiplied by the land area under consideration and is given in m3 s-1 or L s-1. The land area where recharge occurs is called the recharge area or recharge zone. Areas of one to more than hundreds of square kilometers can be involved. A recharge area is particularly vulnerable to any pollutants that may be in the water. Surface sealing, e.g., asphalt parking areas, and buildings, reduces the amount of water that can enter the aquifer from the surface. Groundwater recharge is classified on the basis of the source of water and by the pathway the water takes to enter the saturated zone. Two principal types of recharge are recognized: x Direct recharge is defined as “water added to the groundwater reservoir in excess of soil moisture deficits and evapotranspiration, by direct vertical percolation of precipitation through the unsaturated zone” (SIMMERS, 1997). Infiltration and soil water seepage are the main processes of groundwater recharge. As described in Section 5.2.4, how much water infiltrates depends, among other factors, on climatological aspects,
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
613
vegetation cover, slope of the Earth’s surface, soil composition, depth to the water table, the presence or absence of confining beds. Groundwater recharge depends on vegetation cover and is promoted by flat topography, permeable soil, a deep water table and the absence of confining beds. For groundwater modeling, it is necessary to recognize the variability of recharge within the area under consideration (example in Fig. 6.2-21). Direct recharge is also referred to as diffuse recharge. Diffuse recharge normally occurs over large areas, such as a watershed. Changes in land use, for example deforestation, have an impact on groundwater recharge. x Indirect recharge can occur, for example, as localized recharge from water in natural depressions, ponds, channels and losing streams (influent streams), playas, arroyos and wadis or by interflow, vertical leakage across aquitards and underflow from adjacent aquifers. Losing streams, arroyos, wadis, and ponds play an important role for indirect recharge in arid and semiarid areas. Mountain front recharge takes place as unsaturated and saturated flow in fractured rocks and as infiltration along channels flowing across alluvial fans (STEPHENS, 1996). Recharge from water mains, septic tanks, sewers and drainage ditches in urban areas must also be taken into account for indirect recharge. Local recharge occurs when water infiltrates from perennial and/or ephemeral channels, lakes, etc. and can be modeled as a two- or three-dimensional process. The magnitude of infiltration depends on channel geometry, wetted perimeter, depth and duration of water flow, water temperature, clogging layers on the channel floor, and the hydraulic properties of the unsaturated zone. On the basis of streamflow gauge measurements in seven major channels in the Tucson Basin in southeastern Arizona, BURKHAM (1970) found that about 70 % of the runoff into the ephemeral channels was infiltrated. The channel infiltration along a particular reach can be highly variable in time and space. Irrigation can also contribute to groundwater recharge. In humid climates, recharge mostly occurs by diffuse infiltration and percolation of precipitation and can amount to as much as 20 to more than 50 % of the mean annual precipitation (STEPHENS, 1996). Infiltration and percolation of precipitation does not occur continually over the hydrologic year. It depends on the distribution and intensity of precipitation events, land use, vegetation periods, evapotranspiration, antecedent saturation of soil, etc. In Central Europe groundwater recharge starts between October and December and lasts until March to April. From April to October, the main vegetation period, precipitation contributes little to the recharge (MATTHESS & UBELL, 1983). Groundwater discharge takes place as baseflow to streams (Section 5.2.3). The often repeated statement that there is no direct recharge by rainfall through the unsaturated zone in arid regions is based on a fallacy. In arid and semi-arid climates precipitation occurs as intense summer thunderstorms and
614
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
frontal showers in the winter and spring. In a summary of recharge studies in arid and semi-arid areas, STEPHENS (1996) makes the following statement, together with values for such areas in tabular form: “Regardless of the precipitation and evapotranspiration rates, there is the potential for recharge where the infiltration or redistributed water penetrates below the plant root zone. Winter precipitation, especially in a series of storms, is most likely to contribute to deep soil-water movement and recharge. During the winter storm periods when evapotranspiration is at a minimum, a soil-water budget over the wet period is likely to indicate that infiltrated water would penetrate below the root zone and potentially could become recharge. A few studies illustrate that annual mean precipitation (P) and potential evapotranspiration or gross lake evaporation (E) are not adequate indicators of whether recharge occurs in areas of low annual precipitation.” The volume of groundwater recharge as well as its spatial and time distribution must be known to assess, for example, the capacity of an aquifer for municipal water supply, the feasibility of dewatering a mine pit, the risk of groundwater contamination, the effectiveness of proposed remedial actions, as well as the risk of land subsidence and/or saline water intrusion as undesirable consequences of groundwater withdrawal. Groundwater recharge cannot be determined by direct measurement. It is estimated or determined using indirect methods, either in local (point) investigations or in areal investigations for typical soil areas with different land use (hydrotops, local areas within a catchment area with uniform hydrological conditions) and/or catchment area studies. Common methods are x measurements with soil lysimeters, x water balance estimation, x evaluation of groundwater hydrographs, x stream gauging, x soil-waterflow modeling, x estimations using groundwaterflow models, and x chemical methods. Measurements with soil lysimeters Lysimeters (Section 5.2.2) can be used to estimate the part of precipitation available for groundwater recharge. One problem of this method of recharge estimation is the transfer of point lysimeter data to an area. Other problems of lysimeter measurements are discussed in Section 5.2.2.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
615
Water balance estimation For hydrotop or catchment areas groundwater recharge GWR can be estimated using the groundwater budget Equation (5.2.1): GWR = P - ET - Rd - ǻS,
(5.2.22)
where P is precipitation, Rd is the surface runoff and the groundwater discharged from the area, ET is evaporation and transpiration loss, and ǻS is the change in groundwater storage. The annual groundwater recharge of flat areas without substantial surface runoff can be estimated by a simplified equation: GWR = P – ET.
(5.2.23)
But it must kept in mind that Rd, and ǻS are neglected by this equation. For different hydrotops the following specified estimation methods are used in Germany: x BAGAROV & GLUGLA (GLUGLA & FÜRTIG 1997) x LIEBSCHER & KELLER (1979) x DÖRHÖFER & JOSOPAIT (1980) x RENGER & WESSOLEK (1996). Evaluation of groundwater hydrographs Changes in groundwater level (ǻh) can be monitored using groundwater hydrographs. These changes can be correlated with the precipitation in the same time period. If recharge is to be estimated from groundwater hydrographs, a long time period must be evaluated and the “period of reserve” and the “period of consumption” have to be considered separately. Most important is the qualitative and quantitative correlation between precipitation and the rise of water table in the period of reserve (MATTHESS & UBELL, 1983). The period of reserve is defined as the time when precipitation exceeds evaporation. This is the case, for example, during rainy season in Southeast Asia or in the winter in Europe. In this period a linear correlation between the rise of the water table and the precipitation is observed and can be plotted with the precipitation values on the x-axis and the rise in groundwater level on the y-axis as shown in Fig. 5.2-30. In addition to the amount of precipitation, the rise in groundwater level depends on the amount of evaporation, depth to the groundwater table, specific yield, soil texture, surface runoff, and the change in soil moisture. These factors influence the slope of the best-fit straight line and the axis
616
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
intercept (P1). This line characterizes average conditions. The scatter in the values is due to variations in precipitation. The precipitation can be divided into two parts: P1 and P2. P1 corresponds to that part of the precipitation during which no rise in groundwater is observable, i.e., the water is evaporated, is stored in the upper layer as soil water storage (ǻSW), or becomes direct runoff (Rd). P1 = ET + Rd + ǻSW.
(5.2.24)
The other part of rainfall (P2 = P - P1) is the effective precipitation for groundwater recharge (P2 = Peff = GWR), causing the rise in groundwater table. The steeper the line in the correlation plot of Fig. 5.2-30, the more precipitation is necessary to cause the same rise of groundwater table. The slope D is determined by the specific yield (Sy). tan D
P2 'h
Sy .
(5.2.25)
The correlation line is given by the following equation: P = Sy ǻh + P1.
(5.2.26)
Fig. 5.2-30: Correlation between precipitation and rise of groundwater level with and without groundwater outflow, MATHESS & UBELL (1983)
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
617
If there is no groundwater outflow, the specific yield of the recharged aquifer can be estimated as described above from the slope of the line diagram. If there is outflow, however, groundwater recharge GWR by precipitation (i.e., P2) can be further divided into groundwater storage SG and groundwater outflow RG (see Fig. 5.2-30, solid line): GWR = P2 = Sy 'h = SG + RG
(5.2.27)
It is not easy to determine specific yield. If the specific yield is known, SG and RG can be calculated. Guide values for different materials are given in Table 5.2-9. The specific yield may vary due to inhomogeneities in the ground. A representative value must be found. Table 5.2-9: Specific yield for various materials, BELL (1998) Material
Specific yield [%]
gravel sand dune sand sand and gravel loess silt clay till (silty) till (sandy) sandstone limestone shale
15 - 30 10 - 30 25 - 35 15 - 25 15 - 20 5 - 10 1-5 4-7 12 - 18 5 - 25 0.5 - 10 0.5 - 5
Stream gauging Groundwater discharge is estimated by baseflow separation and by stream hydrograph analysis as described in Section 5.2.3. Soil-waterflow modeling The groundwater recharge rate can be estimated by soil-water flow modeling using measurements of water content and water tension as a function of depth and time. This provides detailed information about the percolation in unsaturated zone and groundwater recharge. But the measurements and calculations are difficult. Estimations using groundwater flow models Groundwaterflow modeling (see Section 5.4.3) starts with a conceptual model (preparation of the model of the aquifer-aquitard system, specification of boundary conditions and hydrological stresses), followed by the design of the numerical model, preparation of model inputs (initial conditions, boundary
618
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
conditions, hydrogeological parameters, and hydrological stresses), model calibration and validation, and the scenario simulation. The calibration of the groundwaterflow model can be used to estimate the groundwater-recharge rate compatible with the measured groundwater data (e.g., level of the groundwater table and hydraulic gradient in the model area). This value of the groundwater-recharge rate can be compared with estimations by other methods. Chemical methods The chloride-profile method or chloride-budget method is frequently used in arid and semiarid areas to determine the water balance in the unsaturated zone (BREDENKAMP et al., 1995; EDMUNDS et al., 1990). But it can also be useful under humid climatic conditions for a general (provisional) groundwater recharge estimate (VOIGT, 1990). It must be kept in mind that the chloride concentration typically decreases with increasing distance inland from the sea. The method is based on the assumption that chloride in the infiltration water originates only from precipitation and is subsequently concentrated by evapotranspiration on the way from the Earth’s surface to the groundwater table. Chloride behaves as a conservative tracer during soil-water transport. The annual groundwater recharge rate is estimated from R
P
Rd C input C gw
where R P Rd Cinput Cgw Kv
P
Rd Kv
,
(5.2.28)
is the groundwater recharge rate [mm a-1], precipitation [mm a-1], direct runoff [mm a-1], chloride concentration in rainwater [mg L-1], chloride concentration in the groundwater [mg L-1], and evapotranspiration coefficient or infiltration coefficient.
The quantity P - Rd is called the effective annual precipitation. For humid conditions in northeastern Germany, SCHLINKER (1969) found the infiltration coefficients for various soil substrates given in Table 5.2-10. Table 5.2-10: Infiltration coefficients Kv for various soil types, SCHLINKER (1969) Soil types
Kv
sand with low content of humus sand with high humus content sand, loamy moraine (till), sandy moraine (till), loamy
4 5 6.7 10 20
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
619
The problem in this method is the determination of chloride input (concentration in rainfall, irrigation water and the amount deposited by dry deposition). This requires either a sampling program, which is expensive, or indirect estimation of the chloride concentration of rainfall, because the chloride concentration in precipitation is different in different weather conditions (VOIGT, 1990). In-situ sampling of isotopes, for example, tritium, chlorine-36, oxygen-18 and deuterium, is also used to estimate groundwater recharge (STEPHENS, 1996; HÖLTING, 1992). The sampling must be carried out below the root zone. It is an advantage that these methods facilitate the integration of hydrologic events over long periods (decades to thousands of years). Sometimes it is possible to investigate the infiltration history on the basis of chemical and isotopic depth profiles (HISCOCK, 2005).
5.2.6 Groundwater Monitoring FLORIAN JENN, CLAUS NITSCHE & HANS-JÜRGEN VOIGT
Groundwater monitoring includes the periodic or continuous measurement of groundwater level and/or of groundwater quality data at selected points. This is done for site characterization, assessment of the situation at the site and prediction of future changes. Groundwater monitoring requires the existence and/or installation of high quality groundwater monitoring wells at suitable locations. For the planning of groundwater monitoring, six questions must be answered (LAUTERBACH & VOIGT, 1993): Why?
Objectives of the monitoring must be defined.
Who?
Who is to conduct the monitoring?
What?
What parameter values must be measured?
Where?
Where are the monitoring wells to be located?
When?
Is the monitoring to be carried out at specified time intervals or are measurements to be made continuously?
How?
What monitoring technique/equipment is to be used?
620
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Monitoring strategies LAUTERBACH & LUCKNER (1999) distinguish between information-oriented monitoring and decision-oriented monitoring. The task of informationoriented monitoring is to evaluate changes in quantity and quality of regional groundwater resources. Several parameters are measured to monitor groundwater quality. The parameters depend on the location of wells in the groundwater catchment area. Most wells are monitored for common groundwater constituents, such as calcium, sodium, and iron. Other wells are monitored primarily for nitrate, chloride, organic solvents and/or gasoline additives. The groundwater level and groundwater quality data can be used for groundwater modeling and predicting groundwater supply. Governmental agencies and/or water supply utilities are responsible for this type of monitoring. Monitoring of groundwater quantity has been conducted for a long time in Europe. The oldest groundwater monitoring network has been in operation since 1845 in France. Most of the networks for groundwater quantity monitoring were installed at the beginning of the 20th century. The average length of records is between 30 and 50 years. Monitoring of groundwater quality has been carried out in most European countries since the 1970s and 1980s. A first overview of “Groundwater Monitoring in Europe” was published by the European Environmental Agency (EEA) in their Topic Report No. 14 in 1996 (KOREIMANN et al., 1996). Examples of the objectives of decision-oriented monitoring are: x detection of the spreading of contamination from different emission sources, e.g., landfills, wastewater plants, industrial sites, mining heaps, agricultural activities, and effectiveness check of remediation measures; x determination of the influence of various technical measures (e.g., pump and treat) on the groundwater table (NILLERT & THIERBACH, 1999; PRZYBILSKI & RECHENBERG, 1999); x scientific studies of processes at work in the unsaturated or saturated zone (AARGAARD et al., 2004); x studies of the distribution of dissolved compounds or colloids in groundwater by tracer experiments (ENGLERT et al., 2000; HERFORT et al., 1998; LE BLANC et al., 1991). A flow chart of the planning of decision-oriented monitoring, e.g., for landfills or sites suspected to be hazardous, is given in Fig. 5.2-31.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
Fig. 5.2-31: Flow chart of decision-oriented monitoring
621
622
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
In contrast to information-oriented monitoring systems, decision-oriented monitoring systems must always be based on an analytical or numerical model of the hydrogeological processes (process model) at the site consisting of a conceptual model of the hydrogeological conditions and a model of the impact, for example, of the source of contamination (see also Section 5.4.4). A scheme for evaluation of the contamination source(s) has been developed (VOIGT, 1990), as well as one for monitoring hydrogeological conditions (VOIGT et al., 2000). The model of the processes must show why monitoring is to be carried out and what has to be investigated. It must also provide an answer to the most important question for decision-oriented monitoring: “Where do the monitoring wells have to be installed?” But, before new wells are drilled, the locations of existing wells must be determined. Recommendations for testing the suitability of existing wells are described later in this section. Next, the “when” and “how” the groundwater samples are to be taken must be determined. State-of-the-art sampling methods and equipment, the determination of monitoring parameters (indicator substances), and sampling intervals (with respect to both time and depth) are given in Section 5.3.1. The first sampling and analysis operation must answer the question “Do the results agree with the process model?” If the results do not match, the model must be recalibrated and if necessary the grid of monitoring wells must be optimized. A prediction of future changes can be given based on the calibrated process model. Current practice of groundwater quality monitoring is to take water samples from observation wells, as well as soil and gas samples at vulnerable and hazardous sites at regular intervals for chemical analysis on site and/or in the laboratory. For sampling, it is necessary to drill and install groundwater observation wells at suitable locations and of sufficient diameter to recover sufficiently large water samples. It is also current practice to use multiparameter probes in wells to monitor several environmental parameters, such as temperature, electrical conductivity, pH, redox potential and oxygen content of the water. In order to investigate the geogenic and anthropogenic background of the groundwater it is necessary to install one or two wells in the inflow area of the landfill or site suspected to be hazardous (reference wells, so-called A-wells, Figs. 5.2-32 and 5.2-33). The location should be selected such that no substances from the landfill or site suspected to be hazardous can migrate to these wells.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
623
Fig. 5.2-32: Groundwater monitoring at landfills and sites suspected to be hazardous, principle schematic for arranging groundwater observation wells in the monitoring zone (view from above), for further notes see text, DÖRHÖFER et al., (1991)
Fig. 5.2-33: Groundwater monitoring at landfills and sites suspected to be hazardous, principle schematic for arranging groundwater observation wells in the monitoring zones (hydrostratigraphic section), for further notes see text, DÖRHÖFER et al. (1991)
624
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
The outflow of the area to be monitored, as depicted in Fig. 5.2-32, is divided into three zones. The size of zone 1 (inner monitoring zone) should be such that the time the groundwater will flow to the margin of this zone from the landfill or site suspected to be hazardous is 200 days. The period of 200 days was selected so that the normal semiannual groundwater sampling provides early detection of changes. To determine the 200-day line, the displacement velocity is calculated and then the flow time. Depending upon the size of the landfill or site suspected to be hazardous, several wells are installed within zone 1. These wells (depicted in Figs. 5.2-32 and 5.2-33) are designated as B-wells (interior monitoring zone) and are crucial in early detection of contamination. They should, therefore, be located as close as possible to the source of contamination (e.g., landfill). Zone 2 (exterior monitoring zone) extends from the 200-day line out to the 2-year line. Several wells may be installed in this zone also (designated C-wells in Figs. 5.2-32 and 5.2-33) to monitor the spreading of substances over a broad area. Additional monitoring wells in a third zone (additional monitoring zone, zone 3) are required only if contaminants can be expected in the groundwater outside of zone 2. The success of the monitoring will ultimately depend on the position and length of the screen installed in the well. If the object under observation is located above a thick aquifer, it is necessary to determine the vertical distribution of contaminants prior to installation of the screens. If, for example, the aquifer is fully penetrated by the observation well and the screen covers the entire thickness of the aquifer, the water sample is a mixed sample from the entire depth range. However, since substances will be entering the monitoring well from only one part of the aquifer, it is possible that the concentration of the substance in the sample will be so diluted that it no longer provides a basis for an assessment of the substance load migrating from the landfill or site suspected to be hazardous. It is difficult to decide on the depths of the screens in the case of landfills with a base seal system, since in this case the substances may enter the aquifer as a dot-like source that cannot be easily localized. The screens should be installed where the substances are expected. In thin aquifers the screens should be installed in segments where the highest permeabilities are present. In thick aquifers the wells should be installed as multiple groundwater observation wells (well clusters with screens at different depths or multi-level wells) to obtain data about the vertical distribution of the contaminant plume. The hydrogeological model, which should already exist for the location, is of major importance when the positions of the well screens are decided. The selection of parameters depends on the composition of the material in the landfill or site suspected to be hazardous and the solubility of the substances. Other factors that should be taken into account are the geogenic and anthropogenic background load in the groundwater, which can be determined from samples taken from the A-well(s) in the inflow area. During
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
625
the orientation phase it is necessary to carry out a comprehensive analysis of the chemical content of the groundwater. Later the extent of the analyses can be reduced. It may be possible to analyze only those substances or key parameters that are particularly obvious in the seepage water from the source or which have particular toxicological relevance. For the following types of waste should be analyzed for the given key parameters: x waste from human settlements: dissolved organic carbon (DOC), chemical oxygen demand (COD), ammonia, potassium, iron, manganese, boron, adsorbed organic halogen (AOX), x building waste: electrical conductivity, ammonia, calcium, chloride, sulfate, AOX. The scope of the parameters as well as the frequency of investigation should be adapted to the determined load. The greatest effort will be associated with contaminated sites without seal systems. In the case of landfills installed according to current engineering practice, the investigations can be restricted to key parameters and longer time intervals. Repeated visits to the observation wells are personnel intensive and hence expensive. Further disadvantages are the uncertainties deriving from the sample collection, transport, and the chemical analysis, in addition to the masses of data produced. The advantages are that numerous substances can be analyzed and that the results of chemical analyses can be used in court. All of these procedures yield point data in space and time. To adequately monitor a site, numerous sampling points in space and time are necessary. A new technological development permits continual in-situ monitoring over long periods of time. The concept is based on a combination of measurements made with multi-parameter probes and optical sensor systems to obtain point data together with an electromagnetic (EM) system to obtain spatial data (Chapter 4.9). To avoid the influence of the system components (probes, optosensors, and EM system) on each other, the three components are activated separately. The multi-parameter probes and optosensors can be installed together in a 3-inch borehole. Transmitters and receivers for the EM system are placed in other boreholes to avoid disturbances from the cables of the probes and optosensors. The cost and uncertainties of monitoring can be reduced by using a permanently installed in-situ system. Such a system reduces the frequency of on-site visits to take samples and the subsequent chemical analysis. The disadvantages of the customary system – uncertainties deriving from the sample collection, transport, and the chemical analysis, in addition to the masses of data produced – are avoided or reduced. In addition, it is possible to react more quickly to the data obtained online. The reliability of the data is increased by the combination of point and spatial data.
626
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Installation of groundwater monitoring wells Groundwater monitoring wells provide direct access to the groundwater system. The monitoring system should preserve the in-situ conditions as well as possible. Therefore, minimum standards are necessary in each of the following categories: x drilling methods, x materials, x installation, and x well development. First, the drilling method must be selected. The decision is based on the requirements on the monitoring well, the hydrogeological conditions, the quantity of water required, the depth and diameter needed, and the budget of the project. The drilling methods are described together with their advantages and disadvantages in Section 5.1.3. The two major drilling procedures used in monitoring well installation are hollow stem augers and rotary drills. The most important decision is whether drilling fluid will be used and if so, what kind. When feasible, drilling procedures that do not introduce water or other liquids into the borehole should be used. If the introduction of drilling fluids cannot be avoided, safety precautions must be used to avoid contamination of the zone to be monitored. Components of a monitoring well are the casing, well screen, centralizers, annular sealant, and gravel pack. During storage, assembly, and installation, all materials used in the construction of a monitoring well must be free of contaminants that could affect the monitoring of the groundwater system. PVC is standard for pipes and well screens. All PVC pipes should be threaded to avoid the volatile organic compounds contained in PVC glues, which could contaminate the system. If non-aqueous-phase liquids (NAPLs) or contaminants of high concentration which react with PVC are to be monitored, HDPE or stainless steel should be used instead of PVC. To insure proper installation of the monitoring well, the following procedure is recommended: The screen interval is selected after the borehole has been drilled and logged. The well screen and the well casing are lowered directly within the hollow stem during a hollow stem auger installation. During a rotary-drilled well installation, potable water must be employed to dilute the drilling mud to ensure ease of installation. After the screen has been positioned and centered in the borehole, the gravel pack is put in place from the base of the well to 20 - 50 cm above the top of the well screen. The gravel pack should consist of material with a grain size on the order of the grain size of the surrounding aquifer. If the filter gravel is very coarse, a counterfilter of finer-grained gravel should complete the filter gravel on either end. Bentonite
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
627
is then placed on top of the filter/counterfilter 2 - 5 m thick to prevent entry of water from the surface or overlying layers into the filter and the aquifer. The remaining borehole is then either completely grouted, or filled with sand, sealing off aquicludes with bentonite (i.e. filling material corresponds to the aquifer-aquiclude structure). On top of the borehole, another bentonite or grout seal (and optionally a concrete slab on the ground) protects against intrusion of rain or surface water. Finally, protective casing and locking well caps must be installed to insure that the monitoring well is protected from vandalism and accidental damage. Well development removes the fine-grained material from both the gravel pack and aquifer material adjacent to the borehole mostly by pumping to improve the hydraulic efficiency of the well. The installation and well development is called borehole completion. Depending on the type of screen installation, monitoring wells are divided into x fully penetrating wells, x multi-screen wells, x “nested” groundwater wells, x cluster or group of monitoring wells, and x special wells with equipment for sampling at preselected depths.
Fig. 5.2-34: Types of monitoring wells, after DVWK-Merkblatt 245 (1997)
628
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
The different types of monitoring wells are shown in Fig. 5.2-34. Until about 1995, fully penetrating or multiple-screen wells were constructed for decisionoriented monitoring in most of West Germany and also in most European countries. Fully penetrating wells have screens installed over the entire thickness of the aquifer. This is done to obtain a hydraulically weighted average water sample of the whole aquifer or, using double (straddle) packers, to obtain a sample from a specified depth. Studies conducted by KALERIS (1992) and BARCZEWSKI et al. (1993) show that it is practically impossible to hydraulically separate the different screened zones or the packer intervals (Fig. 5.2-37). The principal differences in construction and the resulting hydraulic conditions in the different types of wells and their downstream areas are illustrated in Fig. 5.2-35. Figure 5.2-36 shows the hydraulic conditions in and around a fully penetrating groundwater well. In this example, the pump is placed just at the upper edge of the screen. The structure of the aquifer system is identified on the basis of the hydraulic conductivities k1 to k4. The flow resulting from pumping for sampling is denoted by an upward arrow. The figure also shows the velocity distribution. The black arrows indicate layers with high groundwater inflow. Vertical mixing takes place in the annular space of the gravel pack, thus depth-oriented sampling is not possible. Moreover, in the intervals between two groundwater sampling events vertical exchange may take place within the gravel pack, the screen, and the well casing. This shows that fully penetrating wells lead to mixing of groundwater from vertically different conditions in the aquifers and particularly affects the downstream area. The same effect is found in wells with leaky casings. It is possible to carry out sampling at specified depths in aquifers using inline packers and/or multi-packers (BARCZEWSKI et al., 1993; LERNER & TEUTSCH, 1995), however, it does not prevent the cross contamination in the area around the monitoring well (mainly downstream). Figure 5.2-37 shows, for example, the influence of different packer systems on the sampling results in a fully penetrating well. In the opinion of the authors, these systems should be used only for indicating the qualitative layering of groundwater and for installation purposes and to carry out vertical groundwater profiling.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
a) Fully penetrating well
b) Multiple screen well
d) Group of monitoring wells
629
c) Nested groundwater well
e) Special well with depth oriented sampling equipment
Fig. 5.2-35: Hydraulic conditions for different well types
Fig. 5.2-36: Hydraulic conditions in and around a fully penetrating monitoring well during sampling, vf : velocity distribution with depth z
630
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Multiple screen wells are constructed with a single casing and screens at different depths depending on the hydrogeological profile. The different screen sections are separated by grouting seals so that there is no hydraulic connection between the different aquifer layers through the annular space between the casing and the sediments. But inside of the casing there can still be mixing of water from different layers (for example mixing of contaminated water by vertical convection). Accordingly, the use of multiple screens monitoring wells requires inline packer and/or multi-packers to be installed for the entire period of monitoring to prevent hydraulic connection between the individual aquifer layers, avoiding vertical mixing of groundwater of different quality from these aquifers. Studies of the effectiveness of different types of wells (e.g., DEHNERT et al., 2001) also describe the risk of mixing of contaminated water downstream of fully and multiple screen wells. So these types of monitoring wells are not recommended for sampling of groundwater at specific depths. In nested wells, several individual well casings are installed in a borehole and the individual screens (usually 1 to 2 m long) are isolated from one another (HUDAK, 1998). As compared with multiple screen monitoring wells, properly constructed nested monitoring wells have the advantage that the different aquifer layers are not hydraulically connected, therefore, the mixing of water of different quality is avoided. Despite this advantage, the large borehole diameter required for this kind of construction and the disturbances in underground that it may cause during drilling as well as the changed groundwater conditions in the screened depth are the key issues to be considered. These factors frequently result into longer purging times, which can be considerably reduced by installing stationary packer pump systems. A further practical problem in construction of the nested monitoring wells is that of sealing the space between the individual groundwater screens. Since very small annular spaces cannot be always effectively sealed, it demands a lot of effort on the part of the drilling company to meet the specified requirements.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
631
Fig.: 5.2-37: Influence of packer types on sampling results in a fully penetrating well, (a) Influence of a fully penetrating well on the electrical conductivity (resistivity) distribution during sampling, u resistivity from CPT sampling, x resistivity from sampling with double packers. (b) Influence of packer types on sampling results in a fully penetrating well, BARCZEWSKI et al. (1993)
632
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Groups/clusters of monitoring wells are built with a single casing in each borehole and the main aim is to achieve no hydraulic connection between individual aquifer layers. This means that the groundwater tables of every single aquifer layer is independent of the other aquifers. However, due consideration must always be given to disturbance in the underground that may be caused during drilling closely spaced boreholes. The search for improvement in establishing groups of measuring points led to development of a special type of monitoring wells as explained below. In special wells with equipment for sampling at preselected depths, minigauging screens are arranged at different depths and are hydraulically separated by impervious seals. In comparison with the well types already discussed, a minimization of changes in original conditions in the screened depth range is obtained by their direct arrangement in the transient area between the gravel pack and the original subsurface material. The periodic exposure of the uppermost screen to the air is a common problem in monitoring wells. It is a function of the dimensions of the screen and the position of the groundwater table. Such exposure may lead to changes in the condition of groundwater in the exposed screened depth due to longterm biochemical reactions (e.g., clogging with iron precipitate). In special wells with equipment for sampling at specified depths, this effect is counterbalanced owing to the small dimensions of the screen. Screen and sample systems in special wells have non-return valves that additionally help prevent a change in original conditions in the sampling area. This is not the case in the monitoring wells already mentioned. Typical examples of special wells are x multi-level screen (e.g., from the company Pumpen-Boese), x SGM percolation-water and groundwater measurement systems (LUCKNER et al., 1992), x BAT system (TORSTENSSON, 1984), and x multi-screen monitoring wells and measuring point groups having stationary packer pump systems. Suitability tests of existing groundwater monitoring wells A preliminary survey to determine whether there are any monitoring wells that may be suitably integrated in the planned monitoring network is always the first step during planning a monitoring program. Any such existing well must satisfy the following minimum requirements in order to be considered for integration in the monitoring network: x must be supported by documentation (e.g., information about coordinates, documentation of the installation design, lithological profile, hydrogeological monitoring data),
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
633
x must not be damaged or filled, and x must have been in use during the entire last 10 years. To determine whether an existing well can be used as a groundwater monitoring well for quality observation, it is necessary to examine the well on the basis of the criteria given in Table 5.2-11 and check that it satisfies all the criteria. Before selecting any existing groundwater monitoring well, it is of fundamental importance to check whether the screened section of this well is properly placed with respect to the area of observation. The best option is to construct a geohydraulic model coupled with a particle tracking model in order to understand the relationship between the sampled depth in the well and the source of contamination. As a first approximation, the groundwater equipotential lines (isohypses) constructed for the aquifer should be used. This approach only takes advective transport into consideration, i.e. hydrodynamic dispersion and the interaction between the contaminants and aquifer material are neglected. To ensure the proper functioning of a monitoring well (or to ensure that the depth of sampling in the well represents the actual conditions in the aquifer), the well casing must be checked for leakage over the entire length. Borehole television or geophysical borehole methods (Chapter 4.8) can be used to check for leakage of the casing, at the joints of the pipes or in the screen. When that has been done, pumping or infiltration tests can be conducted. Figure 5.2-38 shows the results of standard borehole logging to determine the location and quality of the grouting in a monitoring well. This includes the following logs: x caliper, x natural gamma, x neutron log, x density log, x single point resistivity.
634
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Table 5.2-11: Checklist for determining the suitability of an existing monitoring well I Preselection by evaluation of the available inventory documents
Yes
No
1. Are the well and the screen(s) placed at a depth to be monitored, and is the monitoring well type suited for the investigation objective?
2. Is the casing diameter larger than 2.5 inches and if the screen is at a depth > 50 m, is the diameter larger than 4 inches?
3. Does the location of the clay seal guarantee a good representative groundwater sample without influence from other aquifers penetrated by the well?
4. Do the property rights allow safe access to the well; or, alternatively, is it possible to establish such conditions with reasonable effort?
5. Is the groundwater monitoring well easily accessible, or can present obstacles easily be removed?
6. Does the casing material consist of stainless steel, HDPE or hard PVC, or another comparable plastic?
7. Is the groundwater monitoring well externally intact and does the visual appearance match the inventory data? (Close fit of protective casing to the ground, subsidence, cracks, gaps, etc., deformation of casing, deviation of casing from the perpendicular, leaky or missing cover cap, waste deposition in the vicinity of the well.)
8. Does the sounded depth match the stated depth within 0.5 m and is it possible to insert and remove a dummy sampler (e.g., a dummy pump) without problems?
9. Is the screen, after pumping it clean, sufficiently hydraulically connected to the aquifer? (necessary for pump or infiltration testing)
10. Can the tightness of the casing be confirmed by performing a water pressure test in the casing after installing a packer above the screen?
11. Can the position and effectiveness of the clay seals be confirmed (e.g., by geophysical GR, GG, or NN logging), and do they correspond to the stated well construction (see item 3 above)? (If necessary, inspection with a camera to check condition and location of screen section.)
12. Was the initial sampling successful? (Checking of the analysis results for representativity of the monitoring well, deviations require geochemical/biological explanation.)
II On-site visual evaluation
III On-site technical evaluation
The monitoring well is suitable for groundwater quality monitoring; provide well ID sheet with basic data and inspection record. The monitoring well is not suitable for groundwater quality monitoring.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
635
Fig. 5.2-38: Standard borehole logging to determine the condition of a monitoring well
A U, t fS mS, fs, nj U-T fG-gS
made ground clayey silt fine sand medium sand, fine-sandy, very silty silt - clay fine gravel - coarse sand
Fig. 5.2-39: Results of FEL, SGL and ASGG.D logs in a monitoring well, for further information see text, BAUMANN & PETZOLD (2005). Logs by Bohrlochmessung - Storkow GmbH.
636
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Single point resistivity in combination with a temperature log is extensively used to detect entry of water from unscreened aquifers into the well. These influences can be also detected by focused electric log (FEL) as shown in Fig. 5.2.-39. The combination of FEL log with segmented gamma-ray log (SGL) and annular space gamma-gamma density scanner (ASGG.D) give a better overview about the state of the well. These scanning logs – unlike the standard methods – help in indicating the quality of grouting seal in the whole annular space around the casing. In addition to the above-mentioned methods, flowmeter measurements can be used to evaluate the hydraulic connection of the well screen to the aquifer and the extent of clogging. Figure 5.2-40 shows an example where only a small part of the screen is connected with the aquifer, because most of it is clogged.
Fig. 5.2-40: Check of the screen function by borehole geophysical methods, BAUMANN & PETZOLD, (2005). Logs by Bohrlochmessung - Storkow GmbH.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
637
The results of borehole geophysical methods are an important part of the characterization of the suitability of an existing well for monitoring. The multiple screen well, shown in Fig. 5.2-41 was characterized as not suitable because the grouting seals between the screened sections failed to prevent a hydraulic connection between the screens.
Fig. 5.2-41: Example of checking a multiple-screen monitoring well by geophysical logging, BAUMANN & PETZOLD (2005). Logs by Bohrlochmessung - Storkow GmbH.
638
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Simplified pumping or infiltration tests are used to determine the connection between the screen and the surrounding aquifer in order to establish the effectiveness of the monitoring well. The water used in these tests should be taken from the aquifer being monitored. The fundamentals of such types of infiltration tests and pumping tests are described in Section 5.2.7.2. These tests are not a replacement for more complete geohydraulic pumping tests, but only a rough method for quick estimate of parameter values based on simplified equations for idealized flow conditions. They yield an estimate of permeability (k) at a single point, since only a very limited area around the borehole is hydraulically stressed. These methods work particularly satisfactorily in quick investigations for an initial estimate of parameter values of small areas. A first sampling operation can take place if all tests of the inspection scheme (points 1 to 11 in Table 5.2-11) are positively evaluated and, therefore, representative groundwater samples can be taken from the groundwater monitoring well. If the monitoring well cannot be included in the concerned monitoring net, then it is suggested to try to reconstruct it. Identification sheet of monitoring wells If the well is determined to be suitable for the monitoring grid, then an identification sheet (ID sheet) for the monitoring well should be prepared in cooperation with the governmental agency responsible for such. The sheet must contain the following information: x lithological profile of the borehole, x diagram of the well construction, x results of the well tests, e.g., depth of the pump, pumping duration, pumping rate, drawdown, x operating characteristics, e.g., pH, electrical conductivity, redox potential, oxygen content and temperature, x stamp and signature of the assigned engineer's office and/or laboratory. An example of such an ID sheet is given in Figs. 5.2-42 and 5.2-43.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
Fig. 5.2-42: Example of monitoring well ID sheet: front page
639
640
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-43: Example of monitoring well ID sheet: back page
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
641
Equipment for groundwater monitoring Groundwater monitoring instruments are designed to obtain manual or automatic water-level measurements, water samples and recordings of various water parameter values. Water-level measurements are the principal source of information about the hydrologic stresses acting on an aquifer as well as about groundwater recharge and discharge. Water-level meters are used for manual measurements and water-level monitors are used for automatic continuous measurements. Water-level meters are devices that measure the depth to the water in a well or borehole. It is frequently made of a non-stretch measuring tape with a probe on the end of the tape, which completes an electric circuit when it contacts water. Tapes are marked in mm or 0.01 ft. This kind of water level-meter is also called a sounding light. Hydrostatic water-depth probes, submersible pressure transducers and capacitive water depth probes are used for long-term measurements of water depths in standard ranges of 1, 2, 5, 10 and 20 meters. The resolution of waterlevel measurements depends on the measurement range. The water-depth probes are sometimes combined with temperature probes for the 0 to 60 °C range. Ultrasonic sensors can be used for water-level determinations where instruments cannot be submerged in the liquid, e.g., frozen water or corrosive liquids. The probes can be operated in combination with data loggers and telemetry. Two types of water-level monitors are used: x absolute water-level monitors, and x differential water-level monitors with automatic barometric pressure compensation (also called vented water-level loggers). Storms and elevation changes cause changes in barometric pressure. Since 1 hPa or mbar equals 1 cm of water, considerable changes in water level can be expected due to changes in barometric pressure 4. Absolute water-level monitors measure changes in both water level and barometric pressure. To correctly interpret the data from this type of monitor, an external barometric pressure monitor is required. The true water-level reading is calculated from water-level monitor and external barometricpressure monitor data. The overall accuracy of measurements made using an 4 The mean value of the barometric pressure at sea level is 1013.25 hPa. The barometric
pressure decreases rapidly with altitude according to an exponential function. The decrease near the Earth’s surface is about 1 hPa per 8 m. A diurnal and annual variation as well as a variation due to storm events also occurs. Global extreme values of barometric pressure at sea level are 869.9 hPa and 1085.7 hPa. The largest drop (98 hPa) of barometric pressure within 24 hours was observed within the hurricane Wilma in October 2005 (http://de.wikipedia.org/wiki/Luftdruck).
642
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
absolute water-level device is lower than those made using a differential water-level monitor, because the error of both the water-level monitor and barometric pressure sensor must be considered. Differential water-level monitors have vented cable to automatically compensate for barometric pressure changes. Therefore, direct readings of water-level changes are available in the field and no post-processing is necessary. Interfacemeters measure the depth and thickness of floating or sinking hydrocarbon products (both LNAPLs and DNAPLs) in groundwater with an accuracy of 1 mm or 0.01 ft. A conductivity sensor is used to sense the water-level, an optical infrared refraction sensor is used to detect the hydrocarbon layer. Submersible multi-parameter water-quality sensors are equipped with sensors for elec. conductivity, pH, redox potential, temperature and dissolved oxygen (Section 4.9). Turbidity probes are also used for water quality monitoring. The described monitoring well instruments normally fit inside 2-inch wells. They can also be used in streams, springs, ponds and lakes. Meteorological measurements (e.g., precipitation depth, temperature, air pressure) are necessary for the evaluation and assessment of groundwater monitoring data. These measurements can be also carried out automatically and if necessary telemetrically. Periodic groundwater sampling and analysis is conducted to monitor changes in the concentrations of chemical substances in groundwater. Different sampling intervals are used: Some wells are sampled quarterly and others are sampled annually, depending on the substances that are being monitored. Samples must be representative of in-situ conditions of the aquifer or surface water being monitored. Groundwater sampling and analysis are described in detail in Sections 5.3.1 and 5.3.4. Determination of the direction of groundwater flow The groundwater surface (piezometric surface) and the direction of groundwater flow can be determined from the mapping of water-level elevations in wells drilled into the specific aquifer. At least three groundwater observation wells (called a hydrogeological triangle) are required for this purpose, as shown in Fig. 5.2-44. The measured groundwater levels related to a common datum (mean sea level or a local datum) are plotted with their well position on a diagram made to scale. Then, lines of equal water level are drawn at selected levels between the three or more observation wells. If measurements in more than three groundwater observation wells are available, contour maps can be calculated as described in Section 6.1.1. Due to the anisotropy of aquifers, the contour lines are normally not straight lines as shown in Fig. 5.2-44. But in any case the groundwater flow direction is perpendicular to the contour lines.
643
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
A 40.8
Water-table elevation 40.0
40.0 39.0 39.0 38.0
37.0
38.0
36.0 37.0 35.0
C 36.4
36.0
35.0
B 34.5
Direction of groundwater flow Fig. 5.2-44: Determination of groundwater flow direction using a hydrogeological triangle
5.2.7 Determination of Hydraulic Parameters ULRICH BEIMS, RANJEET NAGARE, CLAUS NITSCHE, MICHAEL PORZIG & HANS-JÜRGEN VOIGT
In terms of hydrogeology, the subsurface consists of x aquifers, x aquitards, and x aquicludes. Aquifers are permeable geological strata that can both store and transmit water in significant quantities. Aquitards are geological units that have a low hydraulic conductivity relative to the units above and below it. Aquicludes are geological units that are incapable of transmitting significant quantities of
644
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
groundwater under ordinary hydraulic gradients; such geologic units have a hydraulic conductivity < 10-9 m s-1. The following zones in an aquifer are distinguished on the basis of the degree of saturation with water: x unsaturated or vadose zone, x capillary fringe, and x saturated zone. The unsaturated zone or vadose zone is the zone between the land surface and the top of the saturated zone, i.e. the water table. In this zone some of the pore spaces are filled with air and some are filled with water. The capillary fringe is that part of the unsaturated zone immediately above the water table which contains water due to capillary rise. It is sometimes used synonymously with capillary space or zone of capillarity. How high the water will rise above the water table as a result of capillarity depends mainly on the size of the interstices (which is determined by the grain size) of the soil or rock. The saturated zone is the zone below the water table where every available space (pores and fractures) is filled with water and the fluid pressure is greater than atmospheric pressure. Aquifers are classified as confined or unconfined. A confined aquifer is bounded above and below by an aquiclude or aquitard (confining bed) and whose piezometric surface is above the base of the confining layer. The water in a well tapping a confined aquifer will rise above the base of the confining layer. Unconfined aquifers (phreatic aquifers) are aquifers in which the surface of the water is at atmospheric pressure. The water in a well tapping an unconfined aquifer will rise only to the level of the groundwater table. The type of the hydrogeological unit affects what hydraulic parameters the occurrence, the amount, and flow of water control. The following parameters are preferred to characterize aquifers, aquitards, and aquicludes: Hydraulic conductivity kf (sometimes denoted as K) is defined as the volume of groundwater that will move within a unit of time under a unit hydraulic gradient through a unit cross-sectional area that is perpendicular to the direction of flow. Hydraulic conductivity is preferred to the term “permeability”. Hydraulic conductivity is the permeability to water only, whereas (intrinsic) permeability is a general property of the matrix only, applicable to different fluids. For a particular soil or rock layer, the hydraulic conductivity may not be the same in the horizontal direction as in the vertical direction. Hydraulic conductivity is defined by Darcy’s law. The hydraulic conductivity depends on the intrinsic permeability of the material and on the degree of water saturation. Saturated hydraulic conductivity, kf, describes water movement through a saturated medium. The values of saturated hydraulic conductivity in soil and rock vary within several orders of magnitude, depending on the material (Table 5.2-12).
645
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
Table 5.2-12: Saturated hydraulic conductivity (kf) for typical fresh groundwater conditions, modified from BEAR (1972) kf (cm s-1) kf (ft d-1) relative permeability aquifer unconsolidated sand & gravel
102 101 100=1 10-1 10-2 10-3 10-4 10-5 10-6 10-7 10-8 5 10 10,000 1,000 100 10 1 0.1 0.01 0.001 0.0001 10-5
pervious
good poor well sorted sand well sorted very fine sand, silt, or sand & gravel loess, loam gravel
unconsolidated clay & organic consolidated rocks
semi-pervious
peat highly fractured rocks
layered clay
oil reservoir fresh rocks sandstone
10-9 10-10 10-6 10-7
impervious none
fat / unweathered clay fresh limestone, dolomite
fresh granite
The unsaturated hydraulic conductivity is a function of the water content of the material. As a material dries out, the connected wet pathways through the media become smaller, the hydraulic conductivity decreasing with lower water content in a very non-linear way. The unsaturated hydraulic conductivity may be a fraction of saturated hydraulic conductivity. This is one of the main complications which arises in studying the vadose zone. Darcy’s law states that the discharge of groundwater Q through a porous medium is proportional to the product of saturated hydraulic conductivity kf, the cross-sectional area of flow A and the hydraulic gradient J normal to that area: Q = kf A J
(5.2.29)
It is assumed that the flow is laminar and steady-state. The hydraulic conductivity kf depends on the internal structure of the medium the water flows through and the viscosity of the water. If the density or viscosity are differing from normal groundwater conditions (e.g. by high temperatures, high salt concentrations, or if considering a different fluid, like oil) it may be necessary to adapt the kf value, based on intrinsic permeability of the matrix and the fluid properties. Porosity is the percentage of open pore space in soil, rock or other materials. It is the ratio of volume of open pores and fractures to the total volume of the material. The porosity is determined from cores or from borehole logs. Effective porosity is the portion of pore space in a saturated permeable material where movement of water takes place. It is almost eqal to total porosity, and greater than drainable porosity/specific yield (SARA, 1993). Transmissivity, is a measure of the capability of the entire thickness of an aquifer to transmit water. Transmissivity T is related to the average hydraulic conductivity kf of the saturated part of aquifer by the relationship: T = kf b,
646
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
where b is the thickness of the saturated part of aquifer. Transmissivity is expressed in units of m2 s-1. When the saturated thickness of an unconfined aquifer changes, the transmissivity also changes. Storativity S is the volume of water that an aquifer releases from or takes into storage per unit surface area of the aquifer per unit change in head. It is also called storage coefficient and is equal to the product of specific storage and aquifer thickness. Storativity is vastly different in unconfined and confined aquifers. In confined aquifers, the change in volume is caused only by elastic compression/expansion of water and rock structure. The corresponding storativity values are on the order of 10-6 to 10-4. In the unconfined case, the filling/draining of pore volume (i.e. specific yield) dominates the value of storativity, which is then on the order of 10-2 to 10-1. In an unconfined aquifer, the storativity therefore is equal to the specific yield. Specific storage is the volume of water released from storage per unit volume of a saturated porous medium per unit decrease in the hydraulic head. It has units of m-1. Specific yield is defined as the ratio of the volume of water that will drain from a porous medium (drainable porosity) by gravity to the volume of the porous medium (see also effective porosity). Field capacity is the amount of water soil can hold against the force of gravity after saturation and runoff (usually 1 to 3 days after rainfall). This water is used by plants during the growing season. In soil science, the term field capacity is used, whereas in hydrogeology the same property is called specific retention (compare specific yield). Infiltration rate is the rate that water enters a soil surface within a unit time interval. The unit is depth of water per unit time, for example mm h-1 or cm s-1. Volumetric water content is the volume of water per volume of soil (i.e. solid material and pores). Figure 5.2-45 gives an overview of hydraulic test methods used in site characterization. Besides these field methods for the determination of hydraulic parameters, laboratory methods can also be used (Section 5.2.7.3). Table 5.2-13 shows which methods can be used for the determination of different hydraulic parameters. Figure 5.2-46 gives an overview of the characteristics of hydraulic test methods and the conditions under which they can be used.
647
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations Table 5.2-13: Methods used for the determination of hydraulic parameters Parameter
Method double-ring Guelph borehole slug/bail pulse pumping laboratory infiltrometer permeameter permeameter test test tests test test test test
hydraulic conductivity
X
X
X
X
transmissivity storativity
X X
X
Fig. 5.2-45: Overview of hydraulic test methods used for site characterization
X X X
648
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-46: Characteristics of hydraulic test methods and their applicability, modified from GARTUNG & NEFF (1999)
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
649
5.2.7.1 Infiltrometer and Permeameter Tests FLORIAN JENN, RANJEET NAGARE, MICHAEL PORZIG & HANS-JUERGEN VOIGT
All hydraulic tests work on the principle of reaction to a change in pressure head. Infiltrometers and permeameters are instruments for determining hydraulic parameters of soil, sediment and rock by measuring the discharge through the material when a known hydraulic head is applied. Infiltrometer and permeameter tests can be subdivided into three groups on the basis of the method of causing a change in the head (Table 5.2-14). Table 5.2-14: Classification of hydraulic tests Method for initial changing the head in the well addition of water at the ground surface gradual continual addition or removal of water in or from borehole sudden addition or removal of water in or from borehole 1)
Test methods x
double ring infiltrometer (DRI)
x x x x x x
Guelph permeameter (GP) constant head borehole permeameter test (BP test) variable head borehole permeameter test (BP test) cone penetration technology (CPT)1) slug/bail test (SB test) pulse test
for details see Section 5.1.4
Test method selection The flow chart in Fig. 5.2-47 can be used to select the appropriate hydraulic test methods on the basis of the hydrogeology of the site. The objectives, the subsurface conditions, and the monetary budget govern the selection of the test method used at a site to be investigated. It is important for the person in charge to know the exact reason for the investigations and whether the tests being used to determine the hydraulic conductivity are suitable. Field tests can be used in combination with laboratory methods.
650
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-47: Flow chart to aid selection of a hydraulic test method on the basis of the hydrogeology of the site
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
651
Test preparation Drilling a borehole is the first step for all of the listed test methods, except for the double ring infiltrometer test. The borehole tests can be conducted either in cased or uncased boreholes. The boreholes have to be close to cylindrical shape in order to obtain reliable results. Hence, it is necessary to select a suitable drilling method (Section 5.1.3). It should be noted that only dry drilling methods may be used. Common problems that arise are x deviation of the borehole from cylindrical shape, x nonuniform borehole walls (as a result of caving) in the case of uncased boreholes, x alteration of the borehole wall due to the construction and operation of the well, which may result in decreased hydraulic conductivity (skin effect). Hydraulic tests conducted in a borehole with skin effect will not reflect the actual hydraulic parameters of the surrounding material (YANG & GATES, 1997). The stability of the borehole wall in uncased boreholes has to be ensured. A temporary casing is highly recommended in unconsolidated rock. In the case of consolidated rock it is recommended that uncased boreholes are used for the hydraulic tests. Test setup and execution Each hydraulic test method is suitable only for certain geological and/or hydrogeological conditions. If the geological and/or hydrogeological conditions are complicated, a combination of test methods may be necessary. It may happen that a test method is found unsuitable during execution. In such a case, it can be necessary to modify or substitute the selected test method by a more appropriate one. The coordinates of each test point must be determined and recorded when the test is set up. Most of the standard test setups include pressure transducers. These instruments can be replaced by a transparent tube with a scale and sealed at the top. The head can be increased to a certain level and the decline directly observed. This kind of setup is particularly suitable for tests at shallow depths (2 - 5 m). In order to compare the results with other tests, groundwater temperature and barometric pressure should be recorded. Modern pressure transducers are able to monitor these parameters. Water added in tests must be from the same groundwater at the location of the well, if possible, in order to avoid geochemical reactions in the aquifer. It is important to repeat the test at least three times in the same depth interval. This ensures a high accuracy of the final results. If the borehole was not developed properly, then the repetitions themselves help to develop the
652
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
borehole and the results improve with each repetition. Moreover, if the discrepancy in the results during the repetitions is too large, then the test must be terminated to find out the reasons for the discrepancies. In addition, two or more different initial heads should be used during testing each well. A series of tests in which the initial head varies by a factor of at least two should be conducted. However, the first and last tests should have the same initial head in order to detect a possible skin effect. General Assumptions If not described otherwise, it should be generally understood that the tests and their evaluation methods are based on the following assumptions: x Darcy’s law is applicable, x flow is assumed to be laminar, and x the aquifer is not affected by external energy (e.g., man-made pressure changes). Further assumptions, especially in case of borehole tests, are that x the aquifer is homogeneous with isotropic properties and is quasi infinite in areal extent, x the aquifer has a uniform thickness, x the aquifer is horizontal, and x the well fully penetrates the aquifer and is open through its entire extent. Possible discrepancies in the final results could be attributed to these assumptions not being fulfilled, since it is seldom possible to assume that all of the above conditions are met in the field during actual testing. Double-ring infiltrometer (DRI) Infiltrometers are simple, officially approved, inexpensive instruments for determining the infiltration rate and hydraulic conductivity of soil or surface sediment layer. Single-ring and double-ring infiltrometers and disc permeameters are commonly used for this purpose. Single- and double-ring infiltrometers only measure flow under ponded (saturated) conditions. Doublering infiltrometer measurements are particularly suitable for relatively uniform fine-grained soils of low plasticity and moderate to low resistance to ring penetration. The measuring range is kf = 10-8 to 5·10-5 m s-1 (GARTUNG & NEFF, 1999). This test method may be carried out at the ground surface on bare soil or with vegetation in place or in a pit with a specified depth depending on the investigation objectives but not below the groundwater table or even below a perched water table.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
653
Single-ring infiltrometer measurements are carried out by driving a single ring into the soil, leaving the top of the ring about ten centimeters above the ground. Water is poured into the ring so that either the amount of water within the ring is always held constant with a Mariotte bottle (see Fig. 5.2-49) (constant head) or the water is allowed to drop with time (falling head). The operator records how much water infiltrates into the soil within a given time period. The rate at which water enters the soil is proportional to the hydraulic conductivity of the soil or sediment layer. A double-ring infiltrometer kit consists of two concentric rings with a lid or other cover, a wood block used as an anvil, one or two Mariotte bottles, sledge hammer, shovel and spade, stopwatch, carpenter's level, steel measuring tape, and a means to determine the location of the test point. The outer ring is, for example, 60 cm in diameter; the inner ring 30 cm. Each Mariotte bottle is calibrated and is graduated from 0 - 3000 mL in 50 mL subdivisions. The volume of water needed to maintain a specified level within specified time intervals is recorded together with the time. When a suitable site has been selected, the coordinates of the site have been determined, and the soil surface prepared, trial pits should be dug in the test area in order to determine the soil type. The outer ring is driven into the soil using a wood block to absorb the blows from the sledgehammer. When the outer ring is in place, the inner ring can be centered inside and driven into the ground the same way. The rings should be inserted into the ground deep enough to prevent the water in the ring from flowing laterally at the surface. A depth of about 5 - 10 cm is usually adequate. Sometimes the rings are set into a bentonite slurry in order to reduce the potential for preferential flow along the ring walls (STEPHENS, 1996). The tops of the outer and the inner rings should be at the same level all the way around. The soil surrounding the rings should be free of significant disturbances. If significant disturbances are observed, the rings should be removed, relocated and reset using a technique that will avoid disturbance of the soil. The purpose of using two rings is to create one-dimensional vertical flow of water from the inner ring, as this simplifies the analysis of data. The outer ring helps force the vertical flow of water from the inner ring into the ground. If water flows in one dimension at a steady rate, the infiltration rate is approximately equal to the saturated hydraulic conductivity. Presaturation of the test site can be helpful to minimize the disturbance of the soil surrounding the rings, to reduce the time required for the test, and make it easier to drive the rings to equal depths. Both rings are then filled with water. The rings should be covered appropriately to prevent evaporation. Mariotte bottles or water-filled columns are used to maintain a constant pool of water in the rings. The outer ring can also be supplied with water manually. It is important to keep the water level equal in both rings. The Mariotte bottle (Fig. 5.2-49) may be replaced by a plastic column with a scale (Fig. 5.2-48a). Figure 5.2-50 shows a computercontrolled test setup.
654
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions (a)
(b)
Fig. 5.2-48: Double ring infiltrometer: (a) principle of the method, (b) rings (photograph courtesy of the Technical University of Dresden)
Fig. 5.2-49: Mariotte bottle supplying water at constant head, e.g., for infiltrometer. If the entrance to the siphon is at the same depth as the bottom of the air inlet, then it will always supply the water at atmospheric pressure and will deliver a flow under constant head H, regardless of the changing height of water within the reservoir.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
655
Fig. 5.2-50: Computer controlled DRI test setup
The water level in the column or the Mariotte bottle of the inner ring is recorded at regular intervals during the test. These measurements are continued until steady-state conditions are reached, i.e., when the water level decrease in the Mariotte bottle or column is constant and the infiltration rate slowly converges to an estimated well-determined linear function. The appropriate reading frequency has to be determined from experience and may be more frequent for soil or sediment with a higher hydraulic conductivity. The steady-state flow rate through the inner ring is used to determine the saturated hydraulic conductivity of the material being tested using Darcy’s law (5.2.29), which can be reformulated for vertical, saturated hydraulic conductivity kf [m s-1] as follows: kf = Qi / S ri2, 3 -1
(5.2.30)
where Qi [m s ] is the steady-state flow rate through the inner ring, assuming that the flow within this ring is essentially one-dimensional downward at a unit hydraulic gradient, and ri [m] is the radius of the inner ring. Sorptivity of the soil/sediment can be determined from data from the unsteady phase at the beginning of the test. Double-ring infiltrometer tests take several minutes to hours depending on soil/sediment type. It should, however, be noted that 100 % saturation is never reached in these tests due to the various processes that occur in the unsaturated zone,
656
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
different soil types, and residual soil air. Experience shows that the measured kf value is 50 - 75 % less than the actual kf value (REYNOLDS & ELRICK, 1986). Some disadvantages associated with DRI method are that the rings are very bulky and heavy to move, and that the water levels in both rings have to be adjusted to be equal only by trial and error, which can be time consuming. Tests may take long time due to the fact that the soil mass enclosed within the rings has to be saturated first before the steady rate is attained. It also requires a flat, undisturbed surface, which sometimes is not available. During the experiment it is sometimes necessary to refill the Mariotte bottles. To do this, the tap has to be turned off and this disrupts the experiment. Guelph permeameter (GP) The Guelph permeameter is a widely used constant-head borehole tool for quick and accurate determination of in-situ hydraulic conductivity, soil sorptivity, and matrix flux potential of all types of soil (REYNOLDS & ELRICK, 1985). It is used for soil/sediment in the hydraulic conductivity range of 10-2 to 10-6 cm s-1. GP measurements are carried out in an uncased 2-inch cylindrical borehole in the unsaturated soil zone from a normal operating depth of 15 - 75 cm to a maximum depth of 315 cm. A Guelph permeameter can also be used to investigate the changing water transmission characteristics of the soil with increasing depth in the same borehole. GP measurements are made in a shallow borehole. The borehole is successively deepened and GP measurements are carried out at the desired depths. Figure 5.2-51 shows the principle of the method. A water reservoir, essentially an in-hole Marriotte bottle constructed of concentric transparent plastic tubes, produces a steady flow which creates a saturated volume in the moist, but unsaturated soil surrounding the borehole. The standard equipment includes a tripod, permeameter, well auger, well development and cleaning tools, collapsible water container, hand vacuum pump, and accessories for extending the depth at which the permeameter can be used, ring attachments that allow ring infiltrometer measurements using 10 - 20 cm diameter rings, and a tension adapter attachment which allows a tension infiltrometer extension for measurements under tensional and very low head conditions.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
657
Fig. 5.2-51: Guelph permeameter, principle of the method, modified from GIAKOUMAKIS & TSAKIRIS (1999)
The Guelph permeameter provides point measurements in a heterogeneous three-dimensional soil environment. Therefore, it is the task of the people in charge to decide on the number of repetitions needed to obtain a representative estimate of the average hydraulic conductivity. This can be judged on the basis of the size of the area, the soil type, degree of heterogeneity, and the objectives of investigation. Soil profile descriptions and soil survey reports greatly enhance the understanding of the data obtained with a GP. The field procedure can be summarized as follows: 1. Drill a borehole of required depth. Make sure that the borehole has been developed properly in order to avoid any discrepancies in the results due to the skin effect. 2. Insert the permeameter into the borehole with the water reservoir fully filled. 3. Start the measurement by raising the air inlet tube out of the outlet.
658
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
4. Set the head H to the desired value by adjusting the height of the air inlet tube. 5. Adjust the rate of decrease of the reservoir level until a steady rate r is attained. 6. The steady-state flow rate Q can be calculated by multiplying the steadystate rate r by the cross-sectional area of the reservoir A. 7. The saturated hydraulic conductivity kf, sorptivity So, and soil matrix flux potential Im can then be estimated as described below. It is important to support the weight of the water reservoir in order to avoid sinking of the permeameter tip. An accidental sinking of the outlet tip results in underestimation of the hydraulic conductivity due to lowering of the head and smaller effective infiltration area (BAGARELLO, 1997). This can be prevented by using a tripod to stabilize the GP. Furthermore, it is recommended to install a temporary borehole casing during measurements in unconsolidated materials. The test data can be analyzed using REYNOLDS’ solution of the RICHARDS equation (REYNOLDS et al., 1985) for steady-state flow Q out of a well: 2S H 2k f C S r 2k f 2S H Im
CQ
(5.2.31)
Rearranging this equation for kf we obtain kf
§ C Q 2S H Im · , ¨ 2 2 ¸ © 2S H C S r ¹
where H r
Im C
(5.2.32)
is the steady-state head of the water in the borehole, well radius, matrix flux potential, a dimensionless proportionality constant dependent on H/r; see Fig. 5.2-52 for values.
Equation (5.2.31) is solved for kf and Im using simultaneous equations or a least-squares approach. The matrix flux potential Im can be calculated using an empirical approach for the hydraulic conductivity-pressure relationship using the formula (ELRICK et al., 1990):
Im
kf
D
where kf
D
,
(5.2.33) saturated hydraulic conductivity, empirical parameter; see Table 5.2-15 for values.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
659
Fig. 5.2-52: Graph for determining the correction factor C for different soils, DURNER (2002) Table 5.2-15: Values for the parameter D depending on soil type, ELRICK & REYNOLDS (1992) Soil type coarse sands, highly structured soils most structured soils and medium sands unstructured fine-textured soils compacted clays
D [m-1] 36 12 4 1
Advantages of the method are that the equipment can be transported, assembled and operated easily by one person. Measurements can be carried out in 0.5 to 2 hours depending on the soil type and requires only about 2.5 L of water. Disadvantages are the low maximum depth of application (315 cm). Furthermore, the initial cost and the maintenance costs could be decisive for the decision of using a GP for investigations. Though not confirmed, it is reported that increasing diameters of the test borehole generate higher kf values. According to REYNOLDS & ELRICK (1985), this can be reflected as an auger or sample size effect (BAGARELLO, 1997). Borehole permeameter (BP) Borehole permeameter tests can be used to determine hydraulic conductivity kf in all types of soil/sediment. This is the oldest and the most widely used method worldwide. BP test methods include a wide range of test designs
660
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
which differ to some extent with respect to theory, procedure, apparatus and methods of data analysis. A BP test can be conducted only at the bottom of a borehole. Hence, it is necessary either to interrupt the drilling at the required depth and conduct the BP test, or to drill a set of temporary test boreholes with different depths. The latter is costly and time-consuming, but allows the correct distribution of kf values to be obtained over a greater lateral extent. The method is recommended for formations with kf values between 10-5 and 10-9 m s-1 (GARTUNG & NEFF, 1999). Figure 5.2-53 gives an overview of different BP test procedures. Two basic techniques are used in borehole permeameter tests: x In the variable head test the rate of rise or fall of head relative to the initial groundwater table is determined as a measure of hydraulic conductivity, and x in the constant head test the rate required to maintain a constant head in the borehole is used to determine hydraulic conductivity. The BP test can be carried out under unsaturated conditions with a constant head or with a falling head, as well as under saturated conditions with a constant head, falling head, or rising head. The constant head BP test is the most common method. Variable head BP tests are normally used in soil/sediment with a relatively low hydraulic conductivity. Tests with constant head and falling head are suitable for both saturated and unsaturated conditions. The rising head test can be conducted only under saturated conditions, i.e. below the groundwater table. This test is supposed to minimize the impact of skin effects. A further variant of this method involves injecting water into a packered borehole test section at a predetermined externally applied additional pressure. The injection rate is adjusted to maintain a predetermined pressure head. The steady-state rate required to maintain the pressure head is then used to calculate the hydraulic conductivity. This test is conducted while maintaining a constant head; therefore, it can be classified as a constant head BP test (HEITFELD & HEITFELD, 1992; TRÖGER & POIER, 2000). Accurate results can be obtained even using low pressure heads. In combination with percussion core drilling, BP tests can be conducted down to depths of around 15 m (SCHREINER & KREYSING, 1998). When rotary or circulation drilling methods are used, this depth is limited only by equipment and financial aspects. Figure 5.2-54 shows the typical test setup for a BP test with and without packers. Packers are especially recommended for consolidated rocks or to separate different sections within a multi-screen well.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
661
Fig. 5.2-53: Overview of suitable BP test procedures classified according to site conditions and area of application
662
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-54: Borehole permeameter tests, schematic diagram of the method
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
663
Simple equipment, such as a temporary casing, a water-level meter and a stopwatch are the basic equipment for BP tests. In addition, an electric pump may be required to extract water (for a rise test). To automate data collection, it is recommended to use a pressure transducer with data logger. A winch for lowering and lifting sensors, packers, etc. is desirable for BP tests. Water consumption in these tests depends on the properties of the test sections. Enough water must be supplied during the test to ensure proper results. The temporary casing used in a BP test is usually a plastic tube, which can be reused. The bottom of the well consists of a gravel filter with grain sizes between 2 and 4 mm. In practice, the test setups suggested by the Earth Manual (USBR, 1974) and HEITFELD et al. (1979) are the most commonly used. Figure 5.2-55 shows possible test configurations. Setups (a), (b), and (c) in Fig. 5.2-55 are open bottom-end setups for interrupted drilling. Drilling can be continued after the test is carried out. Setup (a) is suitable for tests in normal borehole casings, percussion core boreholes, or hollow augers. Setups (b) and (c) involve pulling up the borehole casing several centimeters (depending on the thickness of the strata) to create a cavity for conducting the test (Fig. 5.2-55c). In unconsolidated rock, the cavity is filled with gravel to stabilize the borehole (Fig. 5.2-55b). With these three test setups, it is possible to establish an alternating drilling/testing program. In permanent wells, the test setups shown in Fig. 5.2-55d and e are the most common. Usually the screen section is enclosed in a gravel pack. Standardized equipment and the same test setup should be used for each test within a test series to obtain comparable results for kf values. For accurate measurements under both saturated and unsaturated conditions, it is recommended to use the test setup with an open bottom-end screened section (Figs. 5.2-55b, c, and d). The test setup shown in Fig. 5.2-55b can be used in the unsaturated zone, but the cost and the loss of gravel should be considered. It is recommended to seal the space between the borehole walls and the casing above the screen section, preferably by grouting.
664
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-55: Possible BP test configurations
Field procedure for a BP test: 1. Select proper locations for test points on the basis of data as derived from geophysical and borehole investigations. 2. Drill boreholes at the selected locations to the desired depths using the most suitable drilling method. 3. Install a well casing or temporary casing and seal the space between the casing and the borehole walls and at the ground surface to prevent inflow of surface water (Fig. 5.2-54). 4. To ensure proper results, the length of the screen should be no more than 20% of the thickness of the tested stratum (Figs. 5.2-60 and 5.2-61). In open bottom-end tests, the distance from the bottom of the casing to both the top and bottom of the test stratum should be more than ten times the casing radius (Fig. 5.2-62). 5. An existing developed observation well can also be used for a BP test. 6. Measure the static water level in the borehole/well. This should be done only after the oscillations of the water table caused by the installation of the test equipment have completely stopped. 7. There are two procedures for the variable head BP test: falling head and rising head. Water is added (falling head) or extracted (rising head) from the borehole to create a selected head (usually 2 - 5 m). The recovery of the head to the initial level of the water table is then observed by measuring water levels at specified time intervals. The level can then be plotted
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
665
against time. The hydraulic conductivity can be calculated from the resulting curve. This procedure is repeated several times to obtain a reliable average of hydraulic conductivity. The constant head BP test involves creating a head (usually 3 - 5 m) and then adding water to the well at a rate required to keep this head constant, recording the amount added and the time. The rate required to maintain a constant head in the well is a measure of hydraulic conductivity of the test section. 8. The data obtained in the test is analyzed by one of the methods of analysis explained below (Figs. 5.2-56 to 5.2-62). Field procedure using packers: 1. Drill a borehole of the required specifications. 2. Install the test setup as shown in Fig. 5.2-54b. Keep the pressure valve open during installation. Close the valve and fill the test pipe with water until a predetermined pressure is reached. The space above the packer should be continuously checked for leakage. One good way to prevent leakage is to use two packers, above and below the test section. 3. Open the valve and start logging. In constant head tests, vary the rate at which water is pumped into the packered section through the pipe in order to maintain the predetermined pressure. 4. Use the steady-state rate or the change in head to calculate the hydraulic conductivity kf of the stratum. 5. From this pressure-quantitative measurement, the kf value can be obtained. Due to the broad range of possible interpretation methods, one single evaluation procedure recommendation cannot be given. Figures 5.2-56 and 5.2-59 give different equations for calculating kf for different field conditions. The evaluation methods are divided into two parts according to the reference used: HEITFELD (1998) and Earth Manual (USBR, 1974). The following parameters are used in Figs. 5.2-56 to 5.2-62 and in the equations: Q ri ro Tu H L h1 h2 t1 t2
steady-state flow rate of water added, inner radius of casing, radius of borehole, or outer radius of casing, as appropriate, difference between groundwater table and head in casing at time t, constant head in casing, length of screen section, head in casing at time t1, head in casing at time t2, initial time after adding of water, time of reaching equilibrium.
All parameters should be measured in the field.
666
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-56: Recommended evaluation methods according to HEITFELD (1998)
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
667
Fig. 5.2-57: BP test with test section above the groundwater table according to HEITFELD (1998)
Fig. 5.2-58: BP test with test section partially within the groundwater zone according to HEITFELD (1998)
668
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-59: Recommended evaluation methods according to Earth Manual (USBR, 1974), Method E-18
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
669
Fig. 5.2-60: BP test with test section above the groundwater table according to Earth Manual (USBR, 1974)
Fig. 5.2-61: BP test with test section below the groundwater table according to Earth Manual (USBR, 1974)
670
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-62: Open bottom-end BP test according to Earth Manual (USBR, 1974)
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
671
BP tests provide highly accurate results. The technical effort and the cost of carrying out the test are relatively low. BP tests are one of the most established methods for determining the hydraulic conductivity of an aquifer. There are numerous methods available for analysis of BP test data. Except for the rising head method, BP tests are a natural choice for investigating contaminated sites. The considerable time needed for testing formations with a low hydraulic conductivity is the only factor which may prompt a change to other methods, for example pulse tests (see section below). It is important to know the thickness of the strata being tested in advance. This helps to decide the penetration depth of the borehole. A problem may arise when the method proposed in the Earth Manual (USBR, 1974) is used and is mainly associated with the installation of the test section. Slug/bail test (SB test) The SB test is a variable-head single-borehole aquifer test method to determine aquifer parameters such as hydraulic conductivity and storativity. A piece of metal, called a “slug”, with a defined volume is swiftly submerged in a well and the recovery of the water table to its original level is plotted as a function of time (called “slug insertion phase”). After recovery, the slug is quickly removed and the recovery of the water table is again plotted as a function of time. The same effects can also be achieved by adding or bailing a defined volume of water to or from a well. There are several other methods to achieve the effect of sudden addition or removal of specific volume of water to or from a well during a slug/bail test. One recent method uses positive or negative gas pressure (e.g., air or nitrogen) to change the water table level in a well. See FETTER (2001) and BUTLER (1997) for detailed descriptions. The SB test is applicable only under saturated conditions and is recommended for formations with kf values of 10-5 to 10-9 m s-1 (GARTUNG & NEFF, 1999). The water table in cased wells should be above the screened section. In uncased boreholes, flow into the borehole above the water table should be prevented by suitable measures such as packers (consolidated rocks) or a temporary casing. Figure 5.2-63 shows two possible test setups. The packer method (Fig. 5.2-63b) is preferred for cased boreholes with relatively small diameters and several screened sections, but can also be used in uncased boreholes in consolidated rocks. In this kind of setup the test intervals should be separated using single or double packers. A pump is necessary during the extraction phase of the method shown in Fig. 5.2-63b. The generator for supplying electricity and the winch for lowering and lifting tools into and from the well are not depicted in Fig. 5.2-63, but are indispensable.
672
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-63: SB test, setups for the two possible techniques of the method
Fig. 5.2-64: Schematic diagram of a SB test and the resulting pressure as a function of time
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
673
The procedure for slug and bail tests may differ with respect to the method of changing the water level, but they are still the same in principle (Fig. 5.2-64). 1. Drill and develop a borehole to the required depth. 2. Install the test equipment to be used for the selected method. 3. Measure the static water level, i.e., the initial water level. 4. In the first method a metal slug is lowered into the well below the groundwater table. This is termed the “slug insertion phase”. The addition of a known volume of water to the well and the lowering of a slug below the water table level are equivalent. 5. Measure the decrease in water level as a function of time until it once again reaches the initial water level. 6. Similar results can be obtained by removing the slug or withdrawing a known volume of water from the well and measuring the recovery of the water table over time to its initial level. This is termed the “bail phase”. 7. The data are analyzed to determine the desired parameter values. In the second method, a single or double packer is set at the selected depth and the shut-in valve is closed. The test pipe is then filled with water and the valve is opened. The decrease in the water table is measured as a function of time and recorded by a data logger. The slug-withdrawal phase starts by setting the packer at a desired depth. Once the packer is in place, the shut-in valve is closed and water in the pipe is pumped out completely. The valve is opened and the recovery of the water table to its initial level is observed using a pressure transducer. The Bouwer-Rice method (BOUWER & RICE, 1976) is the most widely used method for slug/bail test data analysis. It is simple and yields highly accurate, reproducible results. It is suitable for both fully and partially penetrating wells. Although it was originally developed for unconfined aquifers, it can also be used in confined aquifers if the top of the well screen is in some distance below the bottom of the upper confining layer (FETTER, 2001). Figure 5.2-65 shows the geometry of the borehole for the Bouwer-Rice method, where rc rw L E D
radius of well casing, radius of borehole, length of well screen, distance from the initial water table to the bottom of the test section, distance from the initial water table to the bottom of the tested aquifer, H(t) difference from initial head at time t.
674
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-65: Geometry of the borehole for Bouwer-Rice analysis method, BOUWER & RICE (1976)
The Bouwer-Rice equation to calculate the field-saturated hydraulic conductivity is
kf
§R · rc2 ln ¨ e ¸ © rw ¹ 1 ln § H o · , ¨ ¸ t © Ht ¹ 2L
where Ho Ht Re
(5.2.34)
drawdown at time 0, drawdown at time t, radial distance from the well over which the head is dissipated.
It is not possible to state exactly the value of Re which applies to a particular well. BOUWER & RICE (1976) have also presented a method to estimate the dimensionless ratio Re/rw: when E < D, §R · ln ¨ e ¸ © rw ¹
ª § D E ·º A B ln ¨ « ¸» © rw ¹ » « 1.1 « § E · » L « ln ¨ ¸ » rw «¬ © rw ¹ »¼
1
,
(5.2.35)
675
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
when E = D, 1
§R · ln ¨ e ¸ © rw ¹
ª º « » « 1.1 C rw » . « § E · L » « ln ¨ ¸ » «¬ © rw ¹ »¼
(5.2.36)
A, B, and C are empirical dimensionless parameters which can be obtained from the curve shown in Fig. 5.2-66.
Fig. 5.2-66: Curves for determining the constants A, B, and C, BOUWER & RICE (1976)
The change in water level in the borehole is continuously recorded. This data is plotted semi-logarithmically versus time (Fig. 5.2-67). The values of H0 and Ht can be determined from this curve. Although the terms H0 and Ht are used in the equation, they merely represent drawdown values at two times. The kf value can also be determined by substituting drawdown values at any other corresponding points in time during the recovery period. When it is necessary to obtain very accurate results, the pattern of recovery has to be considered in order to select a suitable method for analysis. The water level may recover to the initial level in a smooth manner. This response is called overdamped (Fig. 5.2-68). An overdamped response is observed mainly in aquifers that contain fine-grained to sandy material (SARA, 2003).
676
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-67: Example of test data, SCHREINER & KREYSING (1998)
Fig. 5.2-68: Example of an overdamped response for an SB test
Another possible recovery pattern, which is observed mainly in highly conductive aquifers such as coarse sand and gravel, is when the water level oscillates about the initial water level with its magnitude of oscillation decreasing until the oscillations completely cease (Fig. 5.2-69). This kind of response is termed underdamped and is less often observed than an overdamped response. Different methods of analysis are used for these conditions. For overdamped response the evaluation method developed by BOUWER & RICE (1976) is recommended. For information about other analysis methods see FETTER (2001).
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
677
Fig. 5.2-69: Example of underdamped response for an SB test
A main advantage of SB tests is the extensive knowledge on this method. It is an easy and quick method which can be easily carried out by two persons. The volume for which the kf value measured by this test is representative and comparable to that of a pumping test (see Fig 5.2-46). A large number of subsurface parameter values have to be collected for evaluation. The distance from the groundwater table to the aquiclude has to be known. The manner in which the water table recovers its initial state (underdamped and overdamped responses) may influence the final results. Pulse test (PT) The pulse test is a modified slug/bail test (SB test) for single boreholes and is highly suitable for determining the hydraulic conductivity of low permeability sediments (kf < 10-8 m s-1) in the saturated zone. The equipment is the same as that for an SB test with packers. PT methods can also be used to determine transmissivity and skin effect. A short pressure pulse is generated in a test interval separated from other parts of the borehole by packers. The interval is then closed by a valve so that the decay of the pressure impulse (overpressure) is possible only through the rock surrounding the well. Despite the low hydraulic conductivity of the tested formation, the test lasts only 15 to 30 minutes. Figure 5.2-70 shows a schematic of a test setup for pulse tests. If the formation consists of consolidated rock, then the test can be conducted without a casing in a test section closed by a single or double packer. A temporary casing is recommended for unconsolidated rock. The packers are filled with compressed air from a compressor or a gas cylinder. After
678
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
installation of the test equipment (including inflating the packers) the valve is closed and the hydrostatic pressure of the groundwater in the test interval is measured. The installation in the borehole must be checked for water tightness. Then the casing above the closed valve is filled with water until the desired pressure head is reached. The pressure head must be less than that which would crack the rock. The required water is taken from a water tank using a pump if necessary. When the desired pressure head (overpressure) is reached, the valve is opened for a short time (< 10 s) and then closed. The pressure decay in the test interval is measured using a pressure transducer. A high sampling rate is necessary, particularly in the initial phase of pressure decay. Figure 5.2-71 shows the course of a single pulse test run.
Fig. 5.2-70: Schematic diagram of a test setup for pulse tests, POIER (1998)
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
679
Fig. 5.2-71: Course of a single-pulse test run, modified from POIER (1998)
Several methods are available for the analysis of pulse test data (POIER, 1998). Owing to the consistency and reproducibility of the test results, the method developed by PERES et al. (1989) is one of the best ones available to analyze PT data. Peres’ equation for determining hydraulic conductivity is kf
1.151 E U Vg , 2Sm L
where L V rw
E U
g m
(5.2.37)
length of test interval [m], volume of test interval (V = S rw2 L) [m3], radius of well [m], compressibility of water [m2 N-1], density of water [kg m-3], acceleration of gravity [m s-2], slope of the fit line [s per log-cycle], see below.
The value of m is determined graphically: First, values I(dH) need to be calculated for plotting the graph:
680 I dH
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
§ H t H 0 H t 1 H 0 · ¨¨ ¸¸ t t 1 t 0 , 2 H t H max © ¹
where H0 Hmax t0 Ht
(5.2.38)
initial pressure head (water table) [m], maximum pressure head [m], time at application of pressure (i.e. for Hmax) [s], observed head at time t [s].
I(dH) versus time t is plotted on semi-logarithmic paper (example in Fig. 5.2-72), and a straight-line is fitted to the data. The slope of this line is the required value of m. For more details see PERES et al. (1989) and POIER (1998). Pulse tests have been successfully used to determine the hydraulic conductivity of formations with kf values as low as 10-16 m s-1 (BREDEHOEFT & PAPADOPULOS, 1980). One disadvantage associated with pulse tests is that it provides hydraulic parameter values only within a volume of 1 m around the well or borehole. This means the test has to be repeated at a large number of test points. The test should be repeated several times in each hole to ensure accuracy of the final results. Fracturing the rock surrounding the borehole is one factor which may affect the final results.
Fig. 5.2-72: Typical plot of I(dH) versus time t, modified from POIER (1998)
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
681
5.2.7.2 Pumping Tests ULRICH BEIMS
Pumping tests are comparatively time-consuming controllable experiments to determine the hydraulic parameters of an aquifer and performance characteristics of wells. Pumping tests work on a very simple principle: Water is pumped from a well tapping the aquifer, called the pumped well, and the discharge from this well and the changes in the water level in it are observed. The changes in water levels in observation wells (also called piezometers) located at known distances from the pumped well are also observed. From the changes in the water levels in the pumped and observation wells, and from the quality and temperature of the groundwater, it is possible to obtain hydrogeological, hydrochemical, and operational information. Pumping tests are the most reliable methods for characterization of aquifers. According to the objectives and the mode of implementation, the following types of pumping tests can be distinguished: x Well tests: Well tests predominately serve to obtain the performance characteristics of a well. Moreover, the geohydraulic parameters as well as the quality of the groundwater can be estimated. The results of a well test form the basis for planning further pumping tests. x Aquifer tests: Aquifer tests are used to determine the hydraulic conductivity and storage characteristics of an aquifer and the adjacent beds. They provide information about the structure of the aquifer, its hydrochemistry, as well as other information for constructing a groundwater model, e.g., geohydraulic boundary conditions or multipleaquifer formations. In combination with other investigation methods (e.g., geophysical methods) the results can be interpreted in more detail and in combination with tracer tests, the groundwater flow parameters can be better understood. x Intermediate pumping tests: Pumping tests during drilling help to decide about the well depth and well design. x Pumping tests for well development: Pumping tests for well development are carried out to optimize well performance by improving the hydraulics in the well surroundings, minimizing skin effects, and ensuring that the well screen is free of any fine materials that could affect well performance. x Operational tests: Before starting a pumping test or setting up a well for production, it is important to know the reaction of the water table to the pumping. This information is obtained in operational tests to optimize pump settings and later well operation.
682
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
x Long-term pumping tests are conducted to check the ecological effects of groundwater withdrawal and to obtain data on the sustained yield, longterm groundwater composition, and the origin of the water. Figure 5.2-73 shows the stages of a well test with three pumping rates and a subsequent aquifer test. The pumping has to be interrupted for a recovery phase and can be carried out in various combinations. The objectives of a pumping test, such as determination of geohydraulic parameters, information about aquifer structure, optimization of well performance, sustained yield and information on the groundwater composition, determine the type of pumping test, its planning, implementation and evaluation. Not all objectives can be achieved with one type of test. When several types of tests have to be carried out, practicability, order, and combination of the tests have to be considered. A pumping test involves the following phases: x Planning: A tentative program has to be established prior to the start of a pumping test. It includes decisions about the pumping regime, well and piezometer design, the duration of the test, the test parameters and measuring intervals, the measuring instruments, required accuracy, etc. It is appropriate to model the pumping test with estimated parameter values in advance. x Implementation: The implementation phase supervision, and documentation of the test.
includes
organization,
x Evaluation: The first step in the evaluation of a pumping test is to develop a hydraulic model on the basis of the obtained data. Subsequently, the required parameters can be determined and the initial model can be verified.
Fig. 5.2-73: Stages of a pumping test
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
683
Planning of pumping tests The type of the test and the geological and hydrogeological conditions determine what needs to be done to prepare for a pumping test. The following items and their timing have to be considered: x all regulations regarding groundwater extraction and infiltration of extracted water back into the ground, x determination of the need for new observation wells, x briefing everyone involved, x safety and protection measures, x briefing the field crew about the measurements and required accuracy, x instruction about continuation/discontinuation of the test in case of emergencies, x ensuring accurate recording of the drawdown and recovery data (data loggers are preferable to manual measurements in the case of frequent measurements), x selection of a suitable pump and control devices for maintaining a constant pumping rate, x function test of the groundwater observation wells by pumping, borehole geophysics methods or, if necessary, filling the well with water to observe its recovery behavior, and x preparation for sample collection and analysis of the pumped water. For aquifer tests and long-term tests, in addition to the above-mentioned items, attention must be paid to the following aspects: x The drawdown, which immediately follows the withdrawal, is characterized by different phases. Therefore, the duration of the pumping test has to be long enough that the parameter values for the individual drawdown phases can be determined accurately, as well as additional information is ensured. x The speed of reaction of the groundwater piezometric surface strongly depends on whether the aquifer is confined or unconfined. This has to be considered when deciding on the pumping duration. x The degree and distribution of inhomogeneity and anisotropy of the subsurface must be considered when the observation points and measuring intervals are specified. A detailed test plan must contain the following information about the pumping test:
684
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
x the pumping rate, x probable duration of pumping and the time for the recovery measurements, x measurement of the water level, including the initial water level, drawdown and recovery, measuring intervals, required accuracy, and tools, x sample collection (chemical, physical, and microbiological), including sampling tool, sampling location, quantity, extraction time and extent of analysis and sample treatment, x accompanying measurements of precipitation, air pressure, temperature as well as the water level and condition of surface water, x special investigations such as pressure measurements in multiple aquifers, geophysical borehole measurements, chemical profile of indicator, and special parameters as well as isotope and tracer measurements, and x evaluation of the test and calculation of the aquifer parameter values. Implementation of pumping tests Well tests are carried out in several stages and the drawdown relative to the initial water level is recorded. The duration of each pumping stage is usually 4 to 24 hours. The shorter time is suitable for small wells in high permeability aquifers, the longer time is suitable for large wells in low permeability aquifers. There should be three to five pumping stages with pumping rates Qi, i = 1, 2, 3 (, 4, 5) uniformly distributed within the selected range of rates. The measuring intervals should be selected according to Table 5.2-16. This sequence is meant to cover the start of the pumping test, any changes in the withdrawal phase and in the recovery phase. If data loggers are used, measurements are usually taken more frequently. The recovery phase that follows the pumping should be measured for at least two thirds of the pumping time. Table 5.2-16: Measuring intervals as a function of elapsed time since starting a pumping stage Elapsed time since start of a pumping stage
Measuring interval
< 10 min 10 to 60 min 1 to 3 h 3 to 5 h >5h
1 min 5 min 10 min 30 min 1h
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
685
Accurate measurements of the water level at suitable intervals are essential in all pumping stages. Water level measurements in wells at a considerable distance (depends on the aquifer parameters) from the pumped well are usually not necessary. The drawdown s for each pumping stage Qi is plotted versus the elapsed time t since the start of pumping with that rate. The relevant drawdown for the Q-s plot is the drawdown at the end of each pumping stage relative to the water level at the beginning of that stage. In the case of repetition of a well test in order to check the current yield, the later test has to be carried out, whenever possible, with the same specifications as the earlier one. Water samples should be taken for chemical analysis at the end of each pumping stage. Aquifer tests provide comprehensive information about the well and the aquifer. The results of an aquifer test are the basis for all subsequent studies of groundwater quantity and quality as well as on watershed areas. Beyond this, aquifer tests provide well-specific data on well storage and skin effect. For the aquifer under investigation, the pumping duration must be long enough that a geohydraulic model is recognizable and verifiable. Pumping between 200 and 400 hours at a constant pumping rate is needed for this purpose. The shorter duration is suitable for porous rocks, since the parameter values can be determined faster, whereas in hard-rock aquifers with distinct inhomogeneities or leakages the longer duration is needed. For remarks about measuring intervals see section on well tests. The recovery measurements should generally take 0.5 to 0.75 the duration of pumping. At least three water samples should be taken for chemical and/or isotope analysis during the test. For the construction of observation wells for aquifer tests, the hydrological and hydrogeological conditions in the test field, as well as the distance from the pumped well are important. The observation wells should be located in different directions and at different distances from the pumped well in order to observe the drawdown in all directions. The observation well network has to be set up such a way that it takes into consideration local features of the investigation area. Examples of such local features are x geological boundaries, e.g., faults or terrace edges, x hydrological boundaries, such as lakes or rivers, x competitive-use structures, like waterworks or ecologically sensitive areas, and x preferential water flow paths, e.g., drainage ditches. The thickness of the aquifer determines the minimum distance between the observation wells and their respective distances to the pumped well. In homogeneous aquifers the observation wells should be located in the drawdown area not too close to the pumped well and not very near the edges of the cone of depression. Generally, it can be stated that the following values can be used as guidance when positioning the observation wells:
686
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
confined groundwater
M < r < 20 M
unconfined groundwater
H < r < 10 H
where M H r
thickness of a confined aquifer, thickness of an unconfined aquifer, distance of observation well from the pumped well.
For the installation of the observation well network, a rough drawdown calculation using estimated geohydraulic parameters can be used. An imperfect well has no influence on the evaluation if the observation well is at a distance r > M or r > H. Large-scale inhomogeneities, as present in karst and jointed rock aquifers, are not always detectable with a dense observation net. To detect and understand leakage effects during an aquifer test, it is necessary to observe the behavior of the adjacent layers. For aquifer tests, the groundwater levels should be measured with high accuracy, which depends on the expected drawdown and will be on the order of 0.1 % of the drawdown or 1 mm. For manual measurements, an accuracy of about 1 cm has to be achieved. With automatic data recording using data loggers, higher accuracies can be achieved even with short measuring intervals. Depending on the objectives of the test, the following parameters have to be recorded: x water levels in the well and in the annular fill, x pumping rate, x hydrochemical parameters, and x sand content/turbidity. Operational tests yield data important for the well specifications and for later well operations. Operational tests result in a better understanding of the typical behavior of a well during the start-up, i.e., the water level fluctuation immediately after the start of pumping. Operational tests also help in deciding the depth of pump placement and its hydraulic program. Intermediate pumping tests are mainly carried out to test the characteristics of different layers during drilling. They can provide information on differences in yield, for example. They are also carried out in the case of sudden changes in hydraulic properties encountered during the drilling, for example, x water level changes, x separation of multiple aquifers, x fracture zones, and x changes in water quality.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
687
In hard rocks intermediale pumping tests are carried out in open boreholes or boreholes with temporary casings, whereas in porous/unconsolidated rocks they are carried out in boreholes with temporary casings. Intermediate pumping tests are essential for the construction of technically perfect wells or adequately functioning observation wells. Pumping tests for well development help to reduce skin effects, to remove any fine material that may be present in the well screens and to improve the well hydraulics. Objectives of well development are x removal of undersized material in filter gravel, x removal of drilling mud residues, x removal of fine-grained material in the aquifer within a short distance of the well, x construction of a stable grain-skeleton in fine-grained aquifers, x washing out of all mobile particles in the joint system of a hard rock aquifer. In general, well development immediately follows well construction, however in consolidated rock it can be done in open holes. Pumping tests for well development serve only this restricted purpose and are of limited use for further evaluation. Long-term pumping tests are used under the following conditions and/or to achieve the following goals: x in the case of difficult hydrogeological conditions where an analytical approach or modeling is not sufficient, x recognition of long-term changes in the groundwater, x investigation of quantitative and qualitative interactions between groundwater and surface water, x to determine the appropriate pumping rate for a production well which will not significantly affect the ecology and water budget of the aquifer system, and x to demonstrate the effects of groundwater withdrawal on the water budget and ecology. A carefully considered measuring grid and a well coordinated investigation program are necessary for successful implementation of a long-term pumping test. Long-term pumping tests have an impact on the groundwater budget and should be properly justified. The documentation of a pumping test should be done in a pumping test report and comprise the following parts:
688
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
x pumped well, x observation wells, x measured values of the pumped well, and x measured values of the observation wells. An example of a form for the Pumping Test Report – Pumped Well is shown in Fig. 5.2-74. It contains general data about the drilling, times of the start, implementation and end of pumping test, as well as data on well design, and turbidity and color of water. The report is supplemented by a sketch of the well design. The “Pumping Test Report – Observation Wells” is similar to “Pumping Test Report – Pumped Well” and contains technical data about all of the observation wells involved in the test. A map is included in the report showing the locations of the observation wells, their location with respect to lithological and other boundaries, receiving streams, etc., and the design of the observation wells, particularly the position of the screen. The “Pumping Test Report – Measured Values of the Pumped Well” contains all the parameter values recorded during the drawdown and recovery phases. An example showing the minimum requirements is shown in Fig. 5.2-75. The parameters are water levels in the wells, discharge rate (which is recorded by e.g., overfall weirs, water meters or inductive flowmeters), and time of measurement, as well as the properties of the groundwater. The “Pumping Test Report – Measured Values of the Observation Wells” is similar to the one for the pumped well. Automatic systems are often used to record the data. These systems record water level, quantity pumped, as well as selected parameters on groundwater quality. Such systems generally consist of a probe for measuring the data, and a storage device. Many systems can transfer the data directly to a PC, which can evaluate the data with special programs. The accuracy of many devices for measuring water level is close to 1 mm, while the measuring frequency can be as low as 1 s.
689
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations Pumping Test Report – Pumped Well
Site: …………………….…………….
Borehole / well no.: ………………………………
Client: …………………..……………
Contract no.:
Drill foreman: …………..
Test supervisor: …………….
Pumping test no.: ……………
Map sheet no.: ………….
Easting: …………………….
Northing: ………………….... m above MSL
Surface elevation: ………….. Measuring point is ………………………………………. Drainage pipes: …….. m
= …… m above / below surf.
Discharge into ………………………………….
Overfall width of flow rate box: …………………………
Slit shape rectangle / triangle
Water meter reading
end: ………………………….
beginning: …………………..
Other discharge measurements:
Pumping test: st
from: …….... to: …………. o’clock
1 pumping interval
from: …….... to: …………. o’clock =
recovery interval
from: …….... to: …………. o’clock =
2nd pumping interval
from: …….... to: …………. o’clock =
recovery interval
from: …….... to: …………. o’clock =
3rd pumping interval
from: …….... to: …………. o’clock =
recovery interval
from: …….... to: …………. o’clock = pumping
Total time
…… hrs …… hrs …… hrs …… hrs …… hrs …… hrs …… hrs …… hrs
recovery
Drilling method: ………………………
Drilling mud additives: …………………………
Water samples: ………………………
(record on sheet “measured values”)
Borehole depth: ……… m below surface
Position of filter: …………… m below surface
Depth of pump: ……… m below surface
Initial water level:………….. m below surface
Turbidity of water (according to DIN 38404) 0 = clear
1 = slightly turbid
2 = very turbid
3 = opaque
2 = strongly colored
(e.g., brownish)
Color of water (according to DIN 38404) 0 = colorless
1 = slightly colored
Well design (borehole and casing diameters, screened sections and type of casing, seals) see back side
Fig. 5.2-74: Pumping Test Report – Pumped Well
date
time
time since pump start
Time information
Site: Pumping test no.:
drawdown m
water level below meas. point
m
Water level
Borehole / well no.: page:
L s-1 m3 h-1
specific discharge measurerate ment value
Amount of water
PS cm-1
electr. cond. pH
Contract no.:
°C
temp.
Pumping Test Report – Measured Values Pumped Well
cm3 L-1
sand turbidity color remarks content
Properties
690 5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-75: Pumping Test Report – Measured Values of the Pumped Well
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
691
Evaluation of pumping tests The evaluation of pumping tests is a classical “black-box” problem, since only the reaction of the aquifer (drawdown and recovery) to an action (withdrawal) is known, while its hydraulic properties and structure are not known. One basic assumption made for the evaluation of the pumping tests is that the groundwater is confined. However, the evaluation methods discussed in the following sections can also be used for unconfined groundwater by introduction of “reduced drawdown”, sr, which corrects for the fact that in unconfined aquifers the cross-sectional area of groundwater flow changes with the water level. The variables s (drawdown) and M (aquifer thickness) in the formulas for confined conditions then have to be substituted by sr and h0, respectively. sr
s
where sr s h0
s2 , 2h0
(5.2.39) reduced drawdown, measured drawdown, initial (static) water level.
Aquifer tests and long-term tests are suitable for parameter value determination and model identification. The long-term pumping tests supply further information. There are three types of evaluation methods: x graphical/analytical, x computer-aided, and x numerical methods. Graphical/analytical evaluation Non-steady-state flow of groundwater during a pumping test is described by the Theis equation, which is rather complex and not solvable analytically. Using “Theis type curves”, pumping tests can be evaluated graphically by matching type curves to the measured pumping test data. Another method to simplify the Theis equation is the one developed by COOPER & JACOB (1946). The Cooper & Jacob solution of Theis’ equation, however, is valid only if certain conditions are met (basically, this solution is applicable for data that is collected after a sufficient amount of time has passed following the beginning of pumping). The Cooper & Jacob approximation uses a semi-logarithmic plot of the measured data with a best-fit straight line. The following sections introduce the different variants of the best-fit straight line procedure. The basic assumptions made are that the aquifer is quasi infinite in areal extent and does not receive any flow across its
692
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
boundaries. This can be regarded as the standard case for pumping test implementation and evaluation. Best-fit straight line method The best-fit straight line method involves x defining the simplest relationships between the dependent and independent parameters, e.g., s = f (t), s = f (r), s = f (t/r2), s = f (t/t'), or s – s' = f (t'), where f (t) means function of t, for explanation of the other symbols see Equation (5.2.40), x plotting the parameter values and finding the best-fit straight line, x calculation of the required parameter values from the best-fit straight line, and x checking of the validity of the approach. In the case of pumping tests with a constant pumping rate (see Fig. 5.2-76, for an example) there are several possibilities for evaluating the data on the basis of Jacob’s equation: s
Q § 2.25 T t · , ln ¨ 2 4S T © r S ¸¹
(5.2.40)
where (in SI units) s drawdown [m], Q pumping rate [m3 s-1], T transmissivity [m2 s-1], t time after start of pumping when s is measured [s], r distance of the drawdown observation well from the pumped well [m], S storage coefficient [dimensionless]. Different possibilities of evaluation are discussed in the following sections. Evaluation with respect to time The relationship between the measured parameters s and t is described by the straight line s
0.183
Q Q 2.25 T lg t 0.183 lg 2 . T T r S
(5.2.41)
When s and t are plotted semi-logarithmically with t on the log-axis, the result approximates a straight line. The transmissivity T can be determined from the best-fit straight line of the graph (Fig. 5.2-72) and from Equation (5.2.41):
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
T
0.183 Q
' lg t 's
.
693 (5.2.42)
Solving Equation (5.2.40) for S and setting t = t0 and s = 0, we obtain S
2.25t 0 T . r2
(5.2.43)
The value of t0 can be obtained by extending the straight line so that it intersects the t-axis (see Fig. 5.2-77). Then the storage coefficient S can be calculated using this equation. The validity of the Cooper & Jacob approximation must then be verified for the permissible margin of error: t G ! 2.3
S 2 r for H < 5%, T
(5.2.44a)
t G ! 5.0
S 2 r for H < 2%, T
(5.2.44b)
t G ! 8.3
S 2 r for H < 1%, T
(5.2.44c)
where tG
H
is the time when the measurements approach the straight line (Fig 5.2-77) and is the margin of error.
Fig.5.2-76: Drawdown and recovery of a pumping test
694
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-77: Evaluation of drawdown with respect to time
Spatial Evaluation The spatial evaluation is based on the following equation: s
0.366
Q lg r C . T
Figure 5.2-78 shows a plot of drawdown s versus distance r. Observation wells P1 to P4 are shown on the r axis at their respective distances from the pumped well. The drawdown distances in the different observation wells at several pumping time stages t1, t2,… are joined by straight lines. The figure also shows the radius Rt of the cone of depression, which is obtained by extending the straight line through the drawdown values for different wells at a single time until it intersects the r axis (i.e., at a point away from the pumped well where s = 0). The condition for extending such a line for a particular time (t6 in this example) is that for the most distant well the condition in Equation (5.2.44c) has to be met (i.e., error margin less than 1 %). When the measured parameter values are plotted as shown in Fig. 5.2-78, transmissivity and storage coefficient can be calculated from the straight line for time ti: ' lg r
T
0.366 Q
S
2.25 t i T . R t2
's
and
(5.2.45) (5.2.46)
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
695
Fig. 5.2-78: Spatial evaluation of drawdown
Spatial Evaluation with respect to time The spatial-temporal evaluation of pumping tests is based on the relationship of s, t and r, expressed as follows: s
0.183
Q t lg C. T r2
The plot in Fig. 5.2-79 shows the drawdown in different wells plotted as a function of t/r2. The data for three observation wells (P1, P2, and P3) is shown in this example. Theoretically, all plotted values should lie on the same curve. If this is not the case, this indicates that at least one assumption of the underlying model is violated (e.g., no rotation-symmetric flow to the well). The values for transmissivity and storage coefficient can be obtained from this plot:
' lg t / r 2
T
0.183 Q
S
§ t · 2.25 T ¨ 2 ¸ . © r ¹0
's
and
(5.2.47) (5.2.48)
696
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-79: Spatial evaluation with respect to time of a pumping test with three observation wells
Recovery analysis Recovery analysis is based on the superposition principle, since the sudden change in the pumping rate when the pump is switched off, cannot be described mathematically. Instead, the assumption is made that the pump continues to withdraw water from the well. At the time when the pump is stopped, the pump in a second, “virtual” well is switched on into which water infiltrates with an identical rate Q. Hence, the sum of the flow rates in the two wells is zero (reflecting the real-world situation that the pump has been switched off), while mathematically a two-well system with constant Q exists (Fig. 5.2-80).
Fig. 5.2-80: Model for evaluating recovery
697
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
Methods for evaluation of recovery For recovery analysis, two more variables need to be introduced: x residual drawdown, which is the difference between the initial water level and those during recovery: s ' = s - h0 and x time after pumping has stopped: t '. There are two evaluation methods: Method 1: This method uses the residual drawdown s' directly, which is described by s'
0.183
Q § 2.25 T t 2.25 T t ' · , lg lg ¨ 2 T© r S r 2 S ¸¹
(5.2.49)
which can be solved as follows: s'
0.183
Q t lg . T t'
(5.2.50)
Transmissivity can be calculated from the best-fit straight line drawn through the observed data, as shown in Fig.5.2-81:
Fig. 5.2-81: Evaluation of recovery (method 1)
698
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
t '. t 0.183 Q 's ' ' lg
T
(5.2.51)
If the best-fit straight line intersects the t/t' axis at values > 1, groundwater inflow prevails; if it intersects at values < 1, groundwater outflow prevails. A disadvantage of this method is that it is generally not possible to determine storativity S from well recovery. Method 2: This method uses the difference between the actual and residual drawdown for evaluation s s'
0.183
Q § 2.25 T t § 2.25 T t 2.25 T t ' · · ¨ lg lg ¨ lg 2 ¸¸ , 2 T© r S r S r 2S ¹ ¹ ©
(5.2.52)
which is solved as follows: s s'
0.183
Q 2.25 T t ' . lg T r 2S
Fig. 5.2-82: Evaluation of recovery of a pumping test (method 2)
(5.2.53)
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
699
After plotting the data as shown in Fig. 5.2-82, transmissivity and storage coefficient can be obtained using ' lg t '
T
0.183 Q
S
2.25tc0 T . r2
' s s '
and
(5.2.54)
(5.2.55)
This method can be used to determine the skin effect by comparison with the drawdown values. The extrapolation of the drawdown phase on the semilogarithmic plot can usually carried out without any problem. Besides these evaluation methods for the normal pumping test, a number of further practical evaluation methods exist: Pumping tests with constant drawdown For a well where a constant drawdown sc is maintained, the following equation can be used for times t > 103 S/T r02: Q t
T sc . § 2.25 t T · 0.183lg ¨ 2 ¸ © r0 S ¹
Fig. 5.2-83: Pumping test evaluation with constant drawdown sc
(5.2.56)
700
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
When the observed parameter values are plotted as shown in Fig. 5.2-83, transmissivity and storage coefficient can be obtained from the following equations: ' lg t
T
0.183
S
2.25t 0 T . r02
' s / Q
and
(5.2.57)
(5.2.58)
Pumping test with a variable pumping rate In the case of a pumping test with several pumping stages, each with a constant rate, Qj, yielding a well hydrograph that can be approximated by a “step function”, and when (tj – tj-1) > 8.3 S/T r2 holds, the following equation describes the drawdown: s
0.183
2.25 t t j 1 T 1 m§ ¨ Q j Q j 1 lg ¦ T j 1 ¨© r2 S
· ¸. ¸ ¹
Fig. 5.2-84: Evaluation of pumping test with variable pumping rate
(5.2.59)
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
701
When the drawdown s is plotted as a function of the auxiliary variable x as shown in Fig. 5.2-84, transmissivity and storage coefficient can be calculated using the following equations: 0.183 / D x and
T lg
S T
lg
(5.2.60)
2.25 C , r2 D x Qm
where Dx Qm
(5.2.61)
is the slope of the best-fit straight line and the pumping rate in the last stage.
Pumping test with groundwater inflow from adjacent layers For a well with inflow from an adjacent semi-permeable layer, a formula that includes a leakage factor B can be used: s
0.366 Q § B· lg ¨ 1.12 ¸ . T r¹ ©
(5.2.62)
When the measured drawdown values are plotted as shown in Fig. 5.2-85, the parameter values can be determined as follows: T
0.366 Q
B
RB . 1.12
' lg r 's
and
Fig. 5.2-85: Evaluation for inflow from adjacent layers
(5.2.63) (5.2.64)
702
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Pumping test in a two-layered aquifer The evaluation of a two-layered aquifer system is shown in Fig. 5.2-86. In this case, pumping takes place in the lower layer of the aquifer (denoted in the equations by the index l). The drawdown analysis shows three phases. In the first phase, only the pumped lower layer reacts. In the second phase, inflow from the upper aquifer layer (index u) begins, while in the third phase, both aquifers react like a single aquifer, reflecting the drainage properties of the upper layer. From the data of phase I, transmissivity and storage coefficient can be calculated: T
Sl
0.183Q
and
DI
2.25T t 0,I r2
(5.2.65)
.
(5.2.66)
Another set of equations is used for Phase III: T
Su
0.183Q
and
D III
2.25T t 0,III r2
(5.2.67)
.
(5.2.68)
For the evaluation of phase II, the auxiliary variable ty,E is introduced: t y,E
T tE S ur 2
.
(5.2.69)
The required value of tß is the time at the intersection of the best-fit straight line of phase II with the best-fit straight line of phase III. If phase III is not distinct enough for a best-fit straight line to be determined, then the auxiliary variable ts,ß can be used instead: t s,E
T ts . S lr 2
(5.2.70)
The value of ts is obtained from the intersection of the best-fit straight line of phase II with the best-fit straight line of phase I. Next, a second variable E is defined for the conductivity values for vertical and horizontal flow. If 4 d ty,E d 100 or 4 d ts,E d 100,
E can be determined by E
0.195 or E t 1.105 y,E
0.195 . t s,1.105 E
(5.2.71)
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
703
Fig. 5.2-86: Evaluation for delayed yield
The k value for vertical flow through the aquifer can then be obtained from kv
3
E mu T r2
,
(5.2.72)
where mu is the thickness of the upper layer. The factor “3” in this equation was obtained hypothetically from numerical calculations and by comparison with the evaluation pattern of the mathematical model of HANTUSH (1964). Pumping test with boundary conditions If boundary conditions are given, the drawdown at the beginning is similar to the drawdown in an aquifer of infinite extent. When the cone of depression reaches the boundary conditions, the drawdown curve bends (see Fig. 5.2-87): (1) boundary condition “river”: upwards bend (2) boundary condition “barrier”: downwards bend Boundary conditions can also be modeled by appropriate fictitious wells (superposition principle). In addition to the evaluation of phase I as usual, the value of T can be obtained from phase II of drawdown curve (1) as follows: T
Q U ln , 2S s r
(5.2.73)
where U is the distance between the fictitious well and an observation well. Accordingly, from phase II of drawdown curve (2), T can be determined using the following equation:
704
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-87: Influence of boundary conditions
T
0.366 Q
' lg t 's
.
(5.2.74)
The effective distance O* of the boundary condition can be calculated from the time at which the curve bends (time tb):
O*
0.75
tb T . S
(5.2.75)
Procedure of type curves When a type curve procedure is used to evaluate the data, the complete mathematical model of the pumping test is used instead of equations that define the relationships between the parameters. It is necessary to do so when the mathematical model is so complicated that the simplifying assumptions used for best-fit straight line methods are not valid, or there is no analytical solution. In this case, type curves have to be obtained by numerical methods. Even in the case of simple pumping test procedures, it can occasionally be necessary to use type curves if the assumptions for the straight-line approach are not satisfied. If the curves have to be selected from a set of curves, it is sometimes possible to determine not only S and T but also additional independent parameters. If both procedures, best-fit straight line and type curves, can be applied, the one which is more suitable for the actual parameters should be used. The following steps are carried out when type curves are used:
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
705
x The well function is plotted on logarithmic graph paper to obtain a curve, called the type curve. x The observed values are then plotted on a transparent sheet of paper of same scale as the type curve. x The transparent sheet is then superimposed on the type curve so that both curves match as closely as possible. x A match point is selected to obtain four values (two curves, each with two coordinates of the match point, see Fig. 5.2-88), which are then used to calculate the required parameter values as explained in the following sections. Wells with constant delivery The Theis well function, y = W(V), is plotted as a function of 1/V on a sheet of graph paper, and the drawdown s as a function of time t is plotted on another sheet. The four values sm, tm, ym, and (1/V)m for the match point (Fig. 5.2-88) are used to calculate the transmissivity and storage coefficient for the aquifer: T
Q ym , 4S s m
(5.2.76)
S
4T t m . r 2 1/ V m
(5.2.77)
Fig. 5.2-88: Pumping test evaluation using the type curve W(V)
706
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Wells with constant delivery and inflow from adjacent layers A set of curves of y = W(V) is plotted as a function of 1/V and E = r/B on logarithmic paper and drawdown s is plotted as a function of time t on another sheet of logarithmic paper. As shown in Fig. 5.2-89, again four values are obtained for the match point. Transmissivity, storage coefficient, and leakage factor can be calculated using the following set of equations: T
Q ym , 4S s m
(5.2.78)
S
4T t m , r 2 1/ V m
(5.2.79)
B
r
Em
.
(5.2.80)
Fig. 5.2-89: Pumping test evaluation using a set of type curves W(V,E)
Computer-aided procedures The procedures described above – best-fit straight line and type curves – are in wide use because they are easy to use. However, they have some important shortcomings, as for example: x Only some of the data, which has often been obtained with great effort, is used for evaluation.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
707
x Often only a temporal or spatial evaluation is done and any possible difference is not analyzed. x From the individual hydraulic values obtained from the spatial, temporal, drawdown, and recovery analyses, one representative value for the investigated part of the aquifer has to be selected (BUSCH et al., 1993). These disadvantages are mostly avoided by numerical search techniques, because all measured values are taken into consideration simultaneously. Method of evaluation One characteristic of these techniques is the iterative solution of the problem. An initial value which is estimated or approximately determined with other methods is used to solve the problem. Then the values for the potential function h, s and/or for the volume flux Q are compared with the observed values. On the basis of this comparison a new set of geohydraulic parameters is chosen that will yield smaller deviations when the problem is solved again. This is done until the deviations do not become significantly smaller. The biggest difficulty is choosing the new set of parameter values on the basis of the deviation. Fig. 5.2-90 shows a flow chart for the procedure.
Fig. 5.2-90: Flow chart for computer-aided iterative determination of parameter values
The aim of this iterative procedure is to find a strategy for choosing a new set of parameter values. The presently used strategies employ a quality function which is determined by the deviation between measured and calculated values. This quality function has to be minimized by the iteration procedure. For example, the sum of the squares of the differences can be used as a quality function:
708 G
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
¦ D i hM,i hC,i i
where hM,i hC,i
Di
2
,
are measured values, calculated values, and arbitrary weighting factors.
The quality function is a function of the parameters that are to be determined. The minimization of the quality function depends greatly on the number of investigated parameters and on the shape of the quality function “peak”. It has to be checked whether the investigated parameters are independent of each other, since only then can an unambiguous set of parameters be determined. In the case of dependent variables (e.g., T, S and O*), the values are preferably determined in the sensitive areas (parts where there is a clearly established dependency), which often results in a stepwise iterative approach. Figure 5.2-91 presents the result of the program PSU2G for the evaluation of a pumping test, showing the parameter values for each iteration.
Iteration Value of quality function 0 229.1206 1 71.67189 2 20.81663 3 7.332969 4 1.254234 5 0.134818 6 0.115335 7 0.076056 8 0.073032 9 0.072904 10 0.072903
Deviation
0.000283 0.000779 0.000826 0.000747 0.001082 0.000943 0.000278 0.000973 0.000343 0.000044 0.000002
T value
0.001 0.001764 0.002482 0.003223 0.003898 0.004837 0.004993 0.005633 0.005888 0.005915 0.005913
Step S value regarding T value 0.0002 0.001 0.0002 0.000846 0.0002 0.001253 0.0002 0.001352 0.0002 0.002198 0.0002 0.002283 0.0002 0.002053 0.0002 0.001319 0.0002 0.001089 0.00005 0.001054 0.0000125 0.001055
Step regarding S value 0.0002 0.0002 0.0002 0.0002 0.0002 0.0002 0.0002 0.0002 0.00005 0.0000125 0.0000031
The desired parameters have been determined! T value: 0.00591288 S value: 0.00105504 Fig. 5.2-91: Output of the program PSU2G for a digital pumping test evaluation with three observation wells
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
709
Numerical procedures All procedures described above are based on an analytical solution of the flow equation. This requires the aquifer-well system being simplified to a high degree. There is the danger that the real system is simplified to such an extent that the results of a graphical-analytical evaluation are dubious. Pumping test simulator WELL89 For this reason the pumping test simulator WELL89 has been developed. It makes it possible to evaluate pumping tests in a complex aquifer and with a complex well design without inadequate approximations. Figure 5.2-92 presents a discrete model for the program WELL89 for a two-dimensional rotation-symmetric numerical solution. It is a finite-volume model in which the hydraulic quantities are computed on the basis of Darcy’s Law and the geometric dimensions of the segments of the flow field. With the pumping test simulator, the following characteristics of the aquifer can be taken into account: x confined or unconfined, x isotropy or anisotropy, x homogeneous or layered, and x capillary zone. It also offers an opportunity to incorporate the well design characteristics, such as x perfect or imperfect well, x varying well diameter, x multiple screened sections, x capacity of the well, x inflow through the bottom of the well, and x screen losses. The aquifer properties are discretized vertically on the basis of the geological structure and the well geometry and horizontally according to the following rule:
Ui
Ui 1 10 0.25 .
(5.2.81)
The node points are located in the center of gravity of the elements, i.e., they are shifted outwards from the geometric center by an amount of
710 'ri
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Ui 1 Ui 2 . 6 Ui 1 Ui
(5.2.82)
The time steps are initially small and increased as follows: 't i
't i 1 10 0.1 .
Both horizontal and vertical transmissivity and storage coefficient components are considered, as well as the specific elastic storage coefficient and the gravity storage coefficient. The model treats an unconfined groundwater table as a mobile boundary. Transmissivity factors are calculated according to the location of the unconfined groundwater table. The described program has been used successfully for the following four tasks of pumping tests: x analysis of the importance of approximations and assumptions on which the different analytic solutions are based, x determination of flow conditions close to the well, x calculation of type curves for aquifer and well conditions for which no analytical solutions exist, and x interactive pumping test evaluation by means of manual data manipulation. Figure 5.2-93 shows some results of the program WELL89.
Fig. 5.2-92: Rotation-symmetric discretization with pumping test simulator WELL89
711
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations Simulation of drawdown after 2184.7 sec = 36.412 min No. of time step: 16 Task: Comparison WELL – Lohmann drawdown in the well (m): 2.000
Radius drawdown in (m)
layer 1 layer 2 layer 3 layer 4 layer 5
0.2 2 2 2 2 2
0.29 0.01 0.51 1.9 1.9 1.9
0.51 0.01 0.51 1.73 1.73 1.73
0.9 0.01 0.51 1.56 1.56 1.56
1.6 0.01 0.5 1.39 1.39 1.39
2.85 0.01 0.48 1.22 1.22 1.22
5.07 0.01 0.44 1.06 1.06 1.06
9.02 0.01 0.38 0.89 0.89 0.89
16.03 0.01 0.38 0.72 0.72 0.72
Radius drawdown in (m)
layer 1 layer 2 layer 3 layer 4 layer 5
28.51 0.01 0.23 0.56 0.56 0.56
50.07 0 0.15 0.39 0.39 0.39
Well inflow layer 1: Well inflow layer 2: Well inflow layer 3: Well inflow layer 4: Well inflow layer 5: Inflow through bottom:
90.16 0 0.09 0.24 0.24 0.24
160.32 0 0.04 0.11 0.11 0.11
285.1 0 0.01 0.03 0.03 0.03
506.98 0 0 0 0 0
901.55 0 0 0 0 0
1603.21 0 0 0 0 0
0 0 0.005506 0.007337 0.005501 0
Fig. 5.2-93: Result of the program WELL89
5.2.7.3 Laboratory Methods FLORIAN JENN, CLAUS NITSCHE & HANS-JUERGEN VOIGT
In addition to field tests, laboratory methods are used to determine physical parameters of soil, sediment, and rock. Mostly soil and unconsolidated sediment samples are tested in laboratory for petrophysical properties. Consolidated rocks are seldom used for such testing because of their heterogeneous distribution of cracks and joints. Methods for taking disturbed
712
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
and undisturbed soil and unconsolidated sediment samples are described in Section 5.3.2.2. Small soil/sediment samples collected in the field are tested in the laboratory under controlled conditions. The advantage of laboratory tests is that the experiments are carried out “under controlled conditions”. Disadvantages are the problems of sample disturbance and representativeness of relatively small samples. Field and laboratory tests supplement and are a check on each other. Described here are laboratory methods for determining grain size distribution, porosity, and hydraulic conductivity of soil and sediments. Grain size distribution Grain size is the most fundamental physical property of soil, sediment and rock. It refers to the physical dimensions of individual particles. Grain-size analysis, also known as particle-size analysis or granulometric analysis, is the most basic sedimentological technique to characterize soil and sediment. Data from grain-size analysis is used to estimate parameters such as porosity and hydraulic conductivity, and to classify sediments. The traditional method of determining the grain size distribution of soil and sediment samples is sieving for the coarse fractions and the pipette method, based on the “Stokes” sedimentation rates, for the fine fractions (sieve-pipette method). Grain-size analysis is now commonly done using hightech laser instruments. Granular material can range from very small colloidal particles (0.005 µm = 5 nm), to clay, silt, sand, and gravel, to boulders (in the meter range). Owing to the large range of grain sizes that are encountered (more than seven orders of magnitude), a logarithmic scale is used to represent the grain-size distribution. Different scales are used in the USA and in Europe. The European scale is based on particle size measured in mm or µm and plotted using base 10 logarithms. The U.S. scale is also based on measurements in mm or µm, but uses base 2 logarithms to obtain so-called I (phi) values. The U.S. scale is related to the European scale as follows: (5.2.83) phi = I = -log2d = - (log10d / log102), where I d
particle size in I units and diameter of particle in mm.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
713
Fig. 5.2-94: Comparison of grain-size classification according to the Atterberg scale of the European standard DIN EN ISO 14688 (right) with the Wentworth scale (left) used in USA (modified by DOEGLAS (1968), who moved the boundary between the silt and clay fractions from 4 µm to 2 µm), after FÜCHTBAUER (1988)
The grain-size classification according to the Atterberg scale of the ISO/European standard is compared with the Wentworth scale used in USA in Fig. 5.2-94. For grain size analysis it is common practice to take soil samples from each meter of drill core or from each lithological unit. The samples can be split using either a micro-splitter, cone and quartering, or random bulk (KRUMBEIN & PETTIJOHN, 1938). The coarser the sediment is, the larger the volume of material that must be analyzed. The sample must contain enough material to give an adequate representation of the largest particle sizes. Approximately 500 g sample is considered to be appropriate for coarser sediment, while 50 g is sufficient for fine-grained sediment. Standard sieving and sedimentation analysis includes x sample preparation, x drying of the sample in a convection oven (105 °C) and weighing, x dispersion, x wet sieving, x sieving the coarse fraction, x sedimentation of the fine fraction and analysis with the pipette technique.
714
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Sample preparation comprises the removal of organic material and salts. The purpose of removing organic matter is that it does not form clastic grains and is, therefore, not to be considered in a grain-size analysis. Depending on the percentage of organic material in the sediment, sample preparation involves adding different volumes of water and 30 % hydrogen peroxide. Hydrogen peroxide is added until the dark color of the organic matter has largely disappeared. Then the sample is washed three times with NaOAc buffer solution (pH = 5) and once with methanol to remove the remaining released cations (JACKSON, 1956). If the sample contains mineralized water, the sample must be washed with distilled water to remove the salts. Drying of the sample in a convection oven and weighing: Samples are dried in convection ovens at 105 ºC (overnight is usually sufficient). The dry samples are then cooled in desiccators. The samples are weighed immediately after removal from the desiccators to minimize their exposure to the moisture in the ambient air (POPPE et al., 1998). Dispersion: The purpose of this step is to break up the aggregates of claysized particles so that they will be analyzed individually and not as aggregates. This procedure consists of mixing the sediment with a dispersing agent such as sodium metaphosphate (PFANNKUCH & PAULSON, 2005). Wet sieving is done to separate the soil sample into coarse (sand and gravel, which are retained on the sieve), and fine (clay and silt) fractions. The procedure involves placing a sample on a 40 Pm (German standards DIN 18196 and DIN 4022) or 62 Pm (U.S. standard ASTM D 2487) sieve. A funnel above a 1000 mL beaker is placed below the sieve to catch everything that passes through the sieve. If the fine fraction amounts to more than 5 % of the sample, then distilled water with a dispersing agent such as sodium metaphosphate is used during wet sieving. The procedure can be terminated when the wash water leaving the funnel is clear and apparently contains no fines. The fine fraction collected in the beaker is used for sedimentation analysis, while the material on the sieve is subjected to a sieve analysis. Sieving the coarse fraction: The grain-size distribution of the coarse fraction is analyzed by sieving through a tower of sieves (Fig. 5.3-95). The size of mesh openings of the sieves is standardized in Europe as follows: 6, 3.15, 2.0, 1.0, 0.63, 0.315, 0.2, 0.1, 0.063 mm.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
715
Table 5.2-17: U.S. standard sieve mesh, modified after PFANNKUCH & PAULSON (2005) U.S. Standard sieve mesh No. Use wire squares 5 6 7 8 10 12 14 16 18 20 25 30 35 40 45 50 60 70 80 100 120 140 170 200 230 270 325 Analyzed by pipette or hydrometer
Millimeters Millimeters phi (I) (fractional) 4096 -12 1024 -10 256 -8 64 -6 16 -4 4 -2 3.36 -1.75 2.83 -1.5 2.38 -1.25 2 -1 1.68 -0.75 1.41 -0.5 1.19 -0.25 1 0 0.84 0.25 0.71 0.5 0.59 0.75 1/2 0.5 1 0.42 1.25 0.35 1.5 0.3 1.75 1/4 0.25 2 0.21 2.25 0.177 2.5 0.149 2.75 1/8 0.125 3 0.105 3.25 0.088 3.5 0.074 3.75 1/16 0.0625 4 0.053 4.25 0.044 4.5 0.037 4.75 1/32 0.031 5 1/64 0.0156 6 1/128 0.0078 7 1/256 0.0039 8 0.002 9 0.00098 10 0.00049 11 0.00024 12 0.00012 13
Wentworth size class boulder (-8 to -12 I) cobble (-5 to -8 I) pebble (-2 to -5 I)
granule (-1 to -2 I)
very coarse sand (0 to -1 I)
coarse sand (1 to 0 I)
medium sand (2 to 1I)
fine sand (3 to 2 I)
very fine sand (4 to 3 I)
coarse silt (5 to 4 I) medium silt (6 to 5 I) fine silt (7 to 6 I) very fine silt (8 to 7 I) clay (> 9 I)
716
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-95: Sieving shaker with a tower of sieves
The U.S. standard for sieve mesh size based on the phi scale is given in Table 5.2-17. Sieves are arranged in descending order of respective mesh size openings with the coarsest sieve at the top. The individual sand and gravel fractions retained on each sieve at the end of sieving are weighed and recorded. The sieve analysis data is normally directly entered into a computer and is analyzed using special software. Sedimentation methods (hydrometer or pipette) to determine the finegrained fraction: The fine-grained fraction consists of silt (particle size 2 - 62 Pm) and clay (particle size less than 2 Pm, particles smaller than 0.1 Pm are colloidal clay). When the particles are too small for the dry sieving method (i.e. < 74 Pm in the U.S. standard or < 63 Pm in the European standard), a hydrometer or the pipette method is used to separate particles in this size range. The sedimentation methods of particle-size analysis are based on Stokes’ law, which relates the settling velocities of the spherical particles to the size (diameter) and density of the particles, and the viscosity of the fluid in which they are suspended. Stokes’ law of settling velocities can be expressed as follows:
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
v
h t
where
U1
U2 G d 2
18K
v
U2 h G t d
U1 K
d
18K v G U1 U2
717
(5.2.84)
settling velocity [cm s-1] density of the fluid [g cm-3] settling depth [cm] gravity constant [cm s-2] settling time [s] equivalent diameter of the particle [cm] density of the particles [g cm-3] viscosity of the fluid (e.g., water) [g cm-1 s-1].
Hydrometer and pipette methods involve preparing a homogeneous suspension of the fine fraction with a dispersing agent (usually 0.5 % sodium hexametaphosphate) in a 1000 mL graduated cylinder and allowing the particles to settle (KRUMBEIN & PETTIJOHN, 1938; FOLK, 1974; DIN 66115). Table 5.2-18 gives the estimated times and corresponding depths to which a particle with certain grain size will settle for several suspension temperatures. Samples are withdrawn from the suspension and the wt. % are determined (in the pipette method) or measurements are made of the suspension (hydrometer analysis) at defined time intervals. Table 5.2-18: Estimated times and corresponding depths to which a particle with a certain grain size will settle according to Stokes law of settling velocities (POPPE et al., 1998) at several suspension temperatures
I Grain size[µm] 4 63 5 31 6 16 7 8 8 4 9 2 10 1 11 0.5 12 0.24 13 0.12
at 20°C Depth [cm] 20 10 10 10 10 5 5 5
Time 20 s 1 m 56 s 7 m 44 s 31 m 2h3m 4h6m 16 h 24 m 62 h 15 m
at 24°C Depth [cm] 20 10 10 10 10 10 5 5 5 5
Time 20 s 1 m 45 s 6 m 58 s 28 m 1 h 51 m 7 h 24 m 14 h 50 m 59 h 20 m 237 h 20 m 949 h
at 32°C Depth Time [cm] 20 20 s 10 1 m 30 s 10 6m 10 23 m 10 1 h 34 m 5 2 h 58 m 5 6 h 11 m 5 24 h 43 m
718
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
A hydrometer is a device used to measure the density or specific gravity of liquids. It consists of a weighted tube that floats vertically and determines the specific gravity of the liquid by comparison of the scale on the tube with the level it floats in the liquid. ASTM hydrometers are graduated by the manufacturer to read in either specific gravity or in grams per liter (g L-1) and are calibrated for a standard temperature of 20 °C (68 °F). After the initial shaking to homogenize the sediment, aliquots are drawn in the pipette method at a time that the last particles of a given size are just settling past the sampling depth. The aliquots are dried and weighed, which gives the weight of the material finer than the given size. Successive aliquots give the weights of silt and clay particles in each fraction. The weights of the fractions are recorded and the size distribution is determined from the weight of the sediment sample. Conventional grain-size analyses of soil/sediment using a hydrometer or pipette are subject to a significant precision error (ca. 25 - 40 %); they can quantify only a limited number of grain-size classes and are time consuming (SPERAZZA et al., 2002). Laser particle size analyzers employ low-angle light scattering based on Mie theory to rapidly measure grain-size distributions. The light is emitted from a neon-helium laser source. The analyzer can determine a full range of particle sizes in as little as quarter-phi increments from 0.5 µm to nearly 4000 µm, i.e., from very fine clay to fine gravel. In conventional particle-size analysis (sieving and pipetting, for example) the most detailed analyses are commonly at one-phi intervals. Laser grain-size analysis takes typically less than 30 seconds per sample for a complete analysis. Time is required to remove cementing and flocculating agents prior to the laser analysis. In terms of laboratory efficiency, laser diffractometry is far superior to the other methods. This technology is perfect for compositionally homogeneous materials. However, the results of a laser diffractometer reading is influenced by the different optical properties and nonsphericity of the constituent particles of natural sediments (SPERAZZA et al., 2002). For example, the nonsphericity of the particles causes an increase in the observed mean diameter. The platy form of the clay particles causes a considerable difference (as much as eight size classes) between pipette and laser measurements. KONERT & VANDENBERGHE, (1997) found that the < 2 Pm grain size determined by the pipette method corresponds to a grain size of 8 Pm as determined by the Laser Particle Sizer. Because the different measurement techniques can provide different results, different methods should be used and the results compared for an evaluation of the material of the suspension. The grain-size distribution of a soil or sediment is represented in a table or as grain-size distribution curve showing the cumulative weight percent of the different fractions. The data is plotted on semilogarithmic graph paper with particle diameter on the logarithmic x-axis and the percent weight of the fractions on y-axis. In Europe, curves of the cumulative percent weight
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
719
retained by the sieves are generally plotted, which means the cumulative curve will start with the weight of the finest fraction, followed by the successive coarser fractions. In the USA curves of the cumulative weight percent passed by the sieve are normally used, in which each successive fraction is finer than the previous one. Table 5.2-19 shows the results (in phi units) of a grain-size analysis together with the data for the two types of grain-size distribution curves (retained and passing curves) (Fig. 5.2-96). Figure 5.2-97 shows some examples of grain-size distribution curves for different substrates. Table 5.2-19: Example of grain-size analysis data from a sediment sample, PFANNKUCH & PAULSON (2005) Grain size Grain size [mm] (I) 0.01 0.063 0.125 0.25 0.354 0.5 0.707 1 1.41 2 2.83 4
6.64 4 3 2 1.5 1 0.5 0 -0.5 -1 -1.5 -2
Weight of size fraction [g] 0.6 0.5 0.6 1.2 1.8 4.7 17.5 172.2 2570.1 152.7 41.2 0
Weight percent 0.02 0.02 0.02 0.04 0.06 0.16 0.59 5.81 86.74 5.15 1.39 0
Cumulative weight % retained 100 99.98 99.96 99.94 99.9 99.84 99.68 99.09 93.28 6.54 1.39 0
Cumulative weight % passed 0 0.02 0.04 0.06 0.1 0.16 0.32 0.91 6.72 93.46 98.61 100
Other statistical parameters can be derived from the determined grain-size distribution: The coefficient of uniformity (U) characterizes how well or poorly sorted the soil/sediment is: U = d60 / d10 ,
(5.3.85)
where d60 and d10 are the grain diameters in mm for which 60 % and 10 % of the sample is finer (Fig. 5.2-98). Small values of U mean the soil or sediment is well sorted. Sorting is classified on the basis of the coefficient of uniformity according to different standards (Table 5.2-20). The coefficient of uniformity plays an important role in the estimation of porosity, hydraulic conductivity and effective grain-size diameter.
720
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-96: The two types of cumulative weight percent curves, PFANNKUCH & PAULSON (2005)
Fig 5.2-97: Grain size distribution curves of different substrates, FECKER & REIK (1996)
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
721
Fig. 5.2-98: Definition of the diameters d10 and d60 for the determination of the coefficient of uniformity Table 5.2-20: Sorting classes of soil/sediment based on the coefficient of uniformity U according to different standards
well sorted moderately sorted poorly sorted very poorly sorted
DIN 1054
DIN 18196
CASAGRANDE (1948)
U6
U 2.5
722
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
The results of grain-size analysis can be used to classify soils and sediments according to Fig. 5.2-94 in the following manner: The soil or sediment is named by the fraction with the highest weight percent in the total sample. It can be further classified by denoting all fractions with a content of more than 30 % as “strongly” and all fractions with a content between 15 % and 30 % as “slightly”. The remaining fractions are ignored for naming. Consider the example with following weight percent fractions according to grain-size analysis: 1.15 % 22.8 % 43.5 % 31.7 % 0.45 % 0.40 %
gravel coarse sand medium sand fine sand silt clay
The sample name is “strongly fine-sandy, slightly coarse-sandy, medium sand”.
Fig. 5.2-99: U.S. Department of Agriculture method for naming soils (soil texture triangle), USDA (1993)
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
723
Another widely used classification system is the triangle of basic soil texture classes of the U.S. Department of Agriculture (Fig. 5.2-99). In this system the percentages of sand, silt and clay particles are used for the classification of the sample. The soil type is determined by plotting the percentages of each of the three soil particle classes in the soil sample on the soil texture triangle. Porosity Soil, sediment and rock are multi-phase systems consisting of a solid part (solid phase) and voids filled with gaseous and/or liquid phases. The primary voids (pores) can be modified by tectonic processes (development of cracks, joints and fissures), dissolution and cementation, to form secondary pore spaces. The primary minerals of the solid phase must be distinguished from the secondary cement minerals (e.g., quartz and carbonate). Therefore, porosity is a complicated parameter (Figs. 5.2-100 and 5.2-101). Porosity depends on the structure and texture, uniformity of grain size (Fig. 5.2-102), grain shape (Fig. 5.2-103), packing density (Fig. 5.2-104) and degree of cementation of the pore and/or joint spaces (Fig. 5.2-105). The porosity n is a fundamental property of soil/sediment/rock that controls the storage and the movement of water in the ground. Porosity is a dimensionless quantity and can be reported either as a decimal fraction or as a percentage. It can be measured in the laboratory.
A: effective pore space C: dead pore space Fig. 5.2-100: Grain texture of a sediment with early cementation
B: cemented pore space F: primary solid phase Fig. 5.2-101: Texture of clay
724
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Fig. 5.2-102: Porosity is influenced by the degree of uniformity of grain size
Fig. 5.2-103: Kinds of grain shape
Fig. 5.2-104: Types of packing and their influence on porosity
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
725
Fig. 5.2-105: Reduction of the pore space by cementation
The total porosity n of a porous medium is the ratio of the pore volume to the total volume of a representative sample. Assuming that the sample is composed of solid, liquid (water), and gas (air) phases, where Vs is the volume of the solid phase, Vl is the volume of the liquid phase, Vg is the volume of the gaseous phase, Vp = Vl + Vg is the volume of the pores, and V = Vs + Vl + Vg is the total volume of the sample, then the total porosity of the soil sample, n, is defined as n
Vl Vg
Vp
Vs Vl V g
V
.
(5.2.86)
The porosity of the soil/sediment/rock is expressed as total porosity, effective porosity, and specific yield (drainable porosity). Porosity ranges of common unconsolidated and consolidated rocks are given in Table 5.2-22. Only part of the water in a rock flows out when a saturated sample is drained under gravity. The volume of drainable pores is called specific yield or drainable porosity (nd). The amount of water that a unit volume of a sample retains after drainage under gravity is called specific retention or field capacity. The sum of specific yield and specific retention is equal to the total porosity (Fig. 5.2-106). In unconfined aquifers the specific yield is equal to the storage coefficient. Tables 5.2-23 and 5.2-24 compare the ranges of total porosity and specific yield for selected rocks.
726
5.2 Methods for Characterizing the Hydrologic and Hydraulic Conditions
Table 5.2-22: Porosity ranges of common unconsolidated and consolidated rocks, after DRISCOLL (1986) Unconsolidated sediments
n [%]
Consolidated rocks
n [%]
clay silt sand gravel sand & gravel mixture glacial till
45 - 55 35 - 50 25 - 40 25 - 40 10 - 35 10 - 25
sandstone limestone/dolomite (original & secondary porosity) shale fractured crystalline rock vesicular basalt dense, solid rock
5 - 30 1 - 20 0 - 10 0 - 10 10 - 50 toluene > o-xylene > p-xylene > ethylbenzene > m-xylene (Fig. 5.3-39). At a BTEX-polluted site, toluene and o-xylene concentrations were observed to decrease more rapidly in the anaerobic zone than ethylbenzene (BORDEN et al., 1995). In a gasoline plume, toluene and ethylbenzene declined most rapidly with distance from the source, followed by m-xylene, p-xylene, o-xylene and benzene (BORDEN et al., 1997). These examples illustrate that the horizontal and vertical distribution of pollutants provides evidence for biodegradation. Another approach is to use distinctive metabolites for documenting in-situ biodegradation. Such metabolites should have the following characteristics: (i) an unequivocal and unique biochemical relationship to their parent compound, (ii) no commercial or industrial use, and (iii) sufficient biological and chemical stability (BELLER et al., 1995). In the case of metabolites that are subject to further degradation, their absence does not indicate that in situ biodegradation does not occur. For chlorinated ethenes, the detection of the metabolites of reductive dechlorination cis-1,2-dichloroethene (cDCE), vinyl chloride (VC) and ethene (Fig. 5.3-41) along the flow path shows that microbiological reductive
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
885
degradation occurs in the field (SCHMIDT et al., 2006a; TIEHM et al., 2006a). FENNELL et al. (2001) investigated a trichloroethene-(TCE)-contaminated aquifer and observed a significant decrease of TCE and an increase of cDCE and VC in parts of the plume. In contrast, aerobic oxidative degradation of the less chlorinated ethenes cannot be proven by contaminant profiles, since the oxidation leads to complete degradation to carbonate and chloride and not to the formation of stable intermediate metabolites. In these cases, changing isotopic signatures of the contaminants in the field indicate that microbial degradation processes occur on site (see below for further explanation). The formation of metabolites can also be observed in hydrocarbon plumes. Under anaerobic conditions, metabolic by-products include benzoic acid, mono- to tri-methyl benzoic acids and other aromatic acids that are structurally related to alkyl benzene precursors, alicyclic acids and low molecular weight aliphatic acids (COZZARELLI et al., 1994 and 1995; BELLER et al., 1995; LEVINE et al., 1997). It has also been shown that succinate derivatives of toluene, xylene and n-alkanes, for example, are metabolites that indicate anaerobic degradation (BELLER, 2002; GIEG & SUFLITA, 2002; GRIEBLER et al., 2004). Because methylation can be the first step in the degradation of some polycyclic aromatic hydrocarbons (PAH), the corresponding methyl succinic acids can be detectable metabolites, too (SAFINOWSKI et al., 2006). Since pollutant concentration data are available at most investigated sites, the contaminant profiles represent readily available evidence of in-situ biodegradation. Hydrochemical and geochemical analysis The biodegradability of groundwater pollutants differs significantly in different redox environments (Tables 5.3-34 and 5.3-35), which makes it necessary to understand the redox conditions in contaminated aquifers in order to predict in situ biodegradation. Redox conditions are deduced from measurements of redox-sensitive hydrochemical/geochemical parameters such as the redox potential (oxidation-reduction potential, ORP) and concentrations of oxygen, nitrate-nitrite, sulfate-sulfide, dissolved iron(II) and manganese(II), methane and hydrogen (CHAPELLE, 2001). Microbial activity affects the redox regime by consumption of terminal electron acceptors (TEAs) and formation of reduced degradation products. At contaminated sites with high levels of biodegradable material, most of the TEAs present upgradient in the plume are rapidly consumed at or near the source, resulting in a typical sequence of redox zones (Fig. 5.3-42). The availability of the different TEAs significantly affects microbial utilization of specific pollutants. Depletion of oxygen, nitrate, and sulfate within a plume relative to their concentrations outside the plume and elevated concentrations of iron(II) and methane produced within the plume can be used
886
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
to roughly estimate the consumption of electron acceptors in groundwater (BORDEN et al., 1995). However, calculation of the mass of degraded contaminant on the basis of electron acceptor consumption and by-product production is hampered by precipitation and abiotic reactions of compounds in the subsurface. For instance, elevated concentrations of iron(II) only reflect reduction of iron(III) if there is minimal sulfate reduction, because generation of sulfide results in the precipitation of iron sulfide. Iron(II) also may precipitate as ferrous carbonate (HERON & CHRISTENSEN, 1995). The actual distribution of iron(II) will be controlled by the chemical content of the groundwater and the sediment type. The sulfide and carbonate precipitates increase the reducing capacity of the sediment. This will dramatically increase the demand for oxygen if stimulation of microbial degradation by the addition of oxygen is considered (HERON & CHRISTENSEN, 1995; HARTOG et al., 2002). Analysis of hydrochemical/geochemical site data provides rough information which microbial degradation processes are possible considering the prevailing redox zonation and the availability of auxiliary substrates and electron acceptors.
Fig. 5.3-42: Characteristic sequence of redox zones in an aquifer contaminated with biodegradable organics (LNAPL = light nonaqueous phase liquid, DNAPL = dense nonaqueous phase liquid, VCH = volatile chlorinated hydrocarbons)
Isotope fractionation Isotopes are atoms of the same element with different numbers of neutrons, resulting in different atomic weights. Stable, nonradioactive isotopes behave slightly different in chemical and biological reactions owing to this difference in mass. Natural as well as synthetically produced compounds consist of a mixture of heavy (at a low percentage) and light (the majority) isotopes. The ratio of heavy to light isotopes is expressed by the isotopic signature G, which
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
887
can be measured by gas chromatography – isotope ratio mass spectrometry (GC-IRMS). The isotopic signatures of chemicals depend on differences in the chemical feedstock and the production process and, therefore, differ for different production periods and different manufacturers. Thus, isotopic signatures can be used to assign the responsibility for contamination, since isotopic signatures are hardly influenced by abiotic processes in the ground, such as sorption (SCHUTH et al., 2003). Molecules consisting of light isotopes (e.g., 12C) are more rapidly biologically degraded than molecules containing heavy isotopes (e.g., 13C). Thus, microbial degradation causes enrichment of the heavy isotope in the remaining substrate (Fig. 5.3-43). Isotope fractionation of carbon (13C/12C), hydrogen (2H/1H) and chlorine (37Cl/35Cl) atoms during contaminant degradation has been observed for a wide range of contaminants (MORASCH et al., 2001; PELZ et al., 2001; SLATER et al., 2001; BARTH et al., 2002; HUNKELER et al., 2002; NUMATA et al., 2002; MANCINI et al., 2003; CHU et al., 2004; NIJENHUIS et al., 2005) and was reviewed by MECKENSTOCK et al. (2004). Determination that isotope fractionation takes place in the aquifer gives strong evidence that microbial degradation occurs and makes it possible to distinguish biodegradation from abiotic natural attenuation (POND et al., 2002; SHERWOOD-LOLLAR et al., 2001; SONG et al., 2002; CHARTRAND et al., 2005). In the case of complete degradation of contaminants (without formation of metabolites) detection of changing isotopic signatures in the field is the only way to verify microbial degradation analytically. The extent of microbial fractionation is expressed by an isotope enrichment factor. These factors differ for different compounds as well as for different reaction conditions (Fig. 5.3-44), enzymes and bacterial species (BLOOM et al., 2000). Quantification of microbial degradation in the field is possible if appropriate enrichment factors are known (GRIEBLER et al., 2004). Specific enrichment factors can be determined in laboratory degradation studies (e.g., using microcosms) under suitable conditions (HIRSCHORN et al., 2004). The wide range of possible enrichment factors in the literature demonstrates that site-specific enrichment factors should be determined in order to realistically quantify microbial natural attenuation in the field.
Fig. 5.3-43: Principle of isotope fractionation
888
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
Fig. 5.3-44: Range of fractionation factors for degradation of different compounds under different conditions (data from MORASCH et al., 2005; MTBE = methyl tertiary butyl ether, TCE = trichloroethene; 1,2-DCA = 1,2-dichloroethane; cDCE = cis-1,2-dichloroethene; VC = vinyl chloride)
Isotope fractionation enables the detection of microbial degradation processes by measurement of isotopic signatures in field samples. Furthermore, isotope fractionation can be used to quantify microbial degradation processes in the field. Microbial community analysis Most-probable-number (MPN) techniques are used to analyze microbial community composition in terms of metabolically active contaminant degraders or to determine the microorganisms that use different electron acceptors. Metabolically active microorganisms are counted in groundwater samples or soil eluates. Samples are diluted in the wells of microplates (Fig. 5.3-45, left). After incubation, the turbid wells, i.e. wells in which microorganisms grew, are counted (Fig. 5.3-45, right). The most probable numbers are determined using appropriate tables or software based on a Poisson distribution.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
889
Fig. 5.3-45: Most-probable-number determination; left: dilution series; right: counting of the turbid or colored wells
Microorganisms with different metabolic physiologies can be detected with selective media, making it possible to distinguish between active organisms in different redox zones. Total aerobic heterotrophs as well as methanotrophic, ammonia-oxidizing and nitrite-oxidizing bacteria can be detected separately. Anaerobic organisms can be classed as nitrate-reducing, nitrate-ammonifying, iron-reducing, sulfate-reducing or methanogenic microorganisms. Specific contaminant-degrading bacteria can be assessed by adding the relevant contaminants as the sole carbon source (STIEBER et al., 1994). This method can easily be adapted to new compounds, even if the biodegrading microorganisms have not been previously characterized. Analysis of the microbial community present at the site provides an estimate of the number of metabolically active contaminant degraders or microorganisms using different electron acceptors. In combination with hydrochemical/geochemical analyses, redox zones relevant to degradation can be identified. In addition, changes in the microbial community composition during active bioremediation can be monitored. Molecular biological approaches Molecular techniques such as 16S-PCR (polymerase chain reaction, see Fig. 5.3-46 for a general scheme of a PCR reaction) can be used for fast detection of microbial groups, single species, or specific enzymes. The detection of specific contaminant degraders is a promising tool for assessing the degradation potential of a site (FENNELL et al., 2001; HOHNSTOCK-ASHE et al., 2001). During a 16S-PCR reaction, specific parts of 16S-rDNA are amplified, allowing their subsequent detection with agarose-gel electrophoresis. A nested PCR approach is used to increase sensitivity: In the first PCR reaction, universal bacterial 16S-rDNA is amplified, augmenting the DNA material serving as template for the second PCR reaction, the amplification of species-specific rDNA regions.
890
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
Fig. 5.3-46: Principle of a polymerase chain reaction (PCR)
PCR techniques allow a more specific site assessment than MPN methods, since specific degraders can be detected. Moreover, PCR is a very fast technique and, therefore, allows analysis of numerous samples for site assessment. On the other hand, 16S-PCR only provides information about the presence but not about the activity of specific microorganisms. Analysis of the mRNA encoding for degrading enzymes can be a means to compensate for this drawback in the future. Microcosm studies Materials placed in laboratory vessels to measure microbial activity are referred to as microcosms (Fig. 5.3-47). A microcosm is defined as a community or other small unit that is representative of a larger unit. The reasoning for microcosms is that by understanding the activity in a small portion of an aquifer, much can be learned about the aquifer as a whole (TIEHM & SCHULZE, 2003). Microbial degradation studies using microcosms containing the autochthonous, site-specific microflora allow an assessment of the degradation processes occurring in the field. Laboratory (HUNT et al., 1997; BJERG et al., 1999; ALTHOFF et al., 2001; FENNELL et al., 2001; AULENTA et al., 2002; TIEHM & SCHULZE, 2003; TIEHM et al., 2006a) and in-situ microcosms (GILLHAM et al., 1990a and 1990b; ACTON & BARKER, 1992; BJERG et al., 1996; BJERG et al., 1999; GEYER et al., 2005) have frequently been used to demonstrate biodegradation of contaminants by indigenous microorganisms under conditions closely resembling ambient conditions.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
891
Redox conditions are maintained by appropriate techniques for sampling (Fig. 5.3-47, left), transport, handling (Fig. 5.3-47, bottom) and storage (Fig. 5.3-47, right) of the aquifer material. It is also possible to intentionally change conditions in the laboratory in order to understand the effect of changing, for example, redox conditions and availability of specific electron acceptors or auxiliary substrates in the field (HE et al., 2002; LYEW et al., 2002; SCHULZE & TIEHM, 2004; AULENTA et al., 2005; SAGNER & TIEHM, 2005). This is of special interest for understanding the impact of technical remediation or enhanced natural attenuation measures on groundwater chemistry. With microcosm studies, substantiated information is gathered about degradable pollutants and metabolites formed by their degradation under different redox conditions. Whereas field techniques, such as measuring contaminant profiles or isotope fractionation, represent sum parameters for assessing in situ biodegradation, microcosm studies are an appropriate tool for identifying degradation processes.
Fig. 5.3-47: Anaerobic treatment of groundwater microcosms; top left: filling of 2-L bottles with groundwater in the field; top middle: microcosm in the laboratory; top right: anaerobic incubation in an anaerobic vessel equipped with an oxygen-consuming reagent; bottom: anaerobic chamber for the anaerobic handling of microcosms
892
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
5.3.6.3 Case Studies Natural attenuation approaches have been applied for virtually all classes of pollutants (landfill leachates, BTEX (benzene, toluene, ethylbenzene, o-, m-, p-xylene), polycyclic aromatic hydrocarbons (PAH), heterocycles, solvents, pharmaceuticals, pesticides, heavy metals, arsenic and a wide range of halogenated and nonhalogenated organic substances). The most widespread pollutant classes BTEX/PAH and chlorinated solvents are taken as examples to illustrate the methods for site assessment and two contaminated sites are presented as detailed case studies (TIEHM et al., 2002b; SCHULZE et al., 2003; TIEHM & SCHULZE, 2003; SCHULZE & TIEHM, 2004; SAGNER & TIEHM, 2005; TIEHM et al., 2005; SCHMIDT et al., 2006a and 2006b; MÜLLER et al., 2006; TIEHM et al., 2006a and 2006b). Tar oil polluted landfill Site description: The Stürmlinger sand pit (Fig. 5.3-48) is located in Karlsruhe, Germany. The aquifer consists of porous sand and gravel with a composition typical of the upper Rhine valley. The groundwater flow rate averages 0.8 m d-1 and the hydraulic conductivity (kf) is about 210-3 m s-1. Between 1925 and 1956, the former sand pit was filled with municipal and industrial waste, including slag, combustion, and gasworks residues. The disposal site covers an area of about 1500 m2. The bottom of the landfill extends to the groundwater table at about 8 m depth. In the northern part of the dump, tar oil was found as nonaqueous-phase liquid in the groundwater fluctuation zone.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
893
Fig. 5.3-48: Stürmlinger sand pit site, locations of multilevel sampling wells
Contaminant profile: After passing through the landfill, the groundwater is heavily contaminated with BTEX (benzene, toluene, ethylbenzene, o-, m-, p-xylene), PAH (polycyclic aromatic hydrocarbons) and heterocycles. Adjacent to the landfill, the maximum concentration of dissolved BTEX, PAH and heterocycles (concentration given for benzofuran exemplarily) reached values of 50 000 µg L-1, 17 000 µg L-1 and 1300 µg L-1, respectively. The maximum contamination was found at depths of 9 m near the landfill and at 13 m in most of the multilevel sampling wells further downgradient. Concentrations of BTEX and PAH decline along the plume (Fig. 5.3-49). In Fig. 5.3-49, pollutant concentrations are plotted as average concentrations over all five sampling depths of the multilevel wells to eliminate flow path deviations and to increase statistical certainty. All contaminants decrease significantly with increasing distance from the disposal site. The xylenes are no longer detectable (detection limit: 1 µg L-1) at a distance of 170 m and at 260 m only benzene is still measured. At this distance, the concentration of naphthalene is below 1 µg L-1, despite the high initial aqueous phase concentration of about 5000 µg L-1.
894
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
Fig. 5.3-49: Stürmlinger sand pit site, benzene, toluene, ethylbenzene, o-, m-, p-xylene (BTEX, top) and polycyclic aromatic hydrocarbons (PAH, bottom) contamination (average concentrations over the entire depth of the well), absolute (left side) and relative C/C0 (right side, semilogarithmic plot) concentrations are shown (NAP = naphthalene, PHE = phenanthrene, FLU = fluorene, PYR = pyrene, ACE = acenaphthene)
Adjacent to the landfill, 55 % of the total mass of the BTEX and PAH was attributed to benzene, 26 % to naphthalene, and 16 % to toluene, ethylbenzene, and the xylenes (Fig. 5.3-50). At 260 m downgradient, 34 % of the contamination was still attributed to benzene, but naphthalene and the alkyl benzenes represented only 2 % of the remaining contaminants. The contribution of acenaphthene increased from 1 % at a distance of 5 m to 57 % after 260 m. Obviously, both the retardation and biodegradation of acenaphthene in the groundwater plume are poor, resulting in an increasing predominance of this compound. Therefore, acenaphthene might be a suitable analytical substance to monitor maximum plume length. The concentration of naphthalene in the aqueous phase decreased more rapidly (four orders of magnitude within 170 m) than the other, more hydrophobic and more sorptive PAH (two to three orders of magnitude), and toluene was degraded faster than the other alkyl benzenes. This finding is one line of evidence of active biodegradation in the anaerobic zones of the plume.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
895
Fig. 5.3-50: Stürmlinger sand pit site: percentage composition of the sum of benzene, toluene, ethylbenzene, and xylenes (BTEX) and polycyclic aromatic hydrocarbons (PAH) in the contaminant plume at 5 m and at 260 m distance from the landfill source (NAP = naphthalene, PHE = phenanthrene, FLU = fluorene, PYR = pyrene, ACE = acenaphthene)
Redox zonation: Microbial degradation of dissolved contaminants from the landfill is reflected in the concentrations of electron acceptors and donors in the groundwater. Several redox zones were detected in the plume in which reduced concentrations of electron acceptors were observed with the simultaneous appearance of the corresponding degradation products (Fig. 5.3-51). Oxygen and nitrate were detected only in the upper part of the aquifer close to the depth of the water table (9 m); 260 m downgradient, these electron acceptors were also present in the deeper horizons (Fig. 5.3-51 A). Elevated iron(II) concentrations in a large area of the plume indicated reduction of iron (Fig. 5.3-51 B). Sulfate reduction was the predominant redox process in the highly contaminated areas of the plume near the disposal site (Fig. 5.3-51 C). Due to the spatial proximity of iron- and sulfate-reducing processes resulting in precipitation of iron sulfide, only low concentrations of iron(II) and sulfide were found. Elevated methane concentrations were observed only in a small area near the landfill (Fig. 5.3-51 D). The redox zones did not have sharp boundaries and transition zones were observed, indicating that the groundwater samples might have represented composite samples of different microenvironments.
896
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
Fig. 5.3-51: Stürmlinger sand pit site, schematic redox zonation in the longitudinal section (groundwater flow is from right to left): (A) aerobic nitrate-reducing zone (O2 t 0.5 mg L-1 and NO3– t 1 mg L-1); (B) iron-reducing zone (Fe(II) t 1 mg L-1); (C) sulfate-reducing zone (S2- t 100 µg L-1); (D) methanogenic zone (CH4 t 100 µg L-1)
Microbial numbers: The numbers of bacteria were determined in sediment and water samples (SCHULZE & TIEHM, 2004). The spatial distribution of bacteria in groundwater from four wells along the centerline of the plume is shown in Fig. 5.3-52 (see Fig. 5.3-48 for well location). Three samples were taken from each well at different depths. High numbers of total heterotrophic and of contaminant-degrading bacteria (up to 106 g-1 dry sediment) were found. Bacteria that use oxygen, nitrate, iron(III), or sulfate as terminal electron acceptors were detected in most of the samples. Most-probable-number (MPN) results indicated that iron(III) and sulfate reduction were occurring in the same areas and are not sharply separated in the field. In addition, higher microbial numbers and elevated ratios of PAH- and BTEX-degrading bacteria to total heterotrophic bacteria in the highly contaminated areas indicate active biodegradation.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
897
Fig. 5.3-52: Stürmlinger sand pit site, aerobic and anaerobic microbial numbers (per g of sediment dry substance) of four wells in the center line at three different depths (BTEX = benzene, toluene, ethylbenzene, o-, m-, p-xylene, PAH = polycyclic aromatic hydrocarbons)
Degradation processes: Results from microcosm studies show different degradation patterns for BTEX, PAH and heterocycles under different redox conditions. Under sulfate-reducing conditions, toluene and ethylbenzene were degraded (Fig. 5.3-53, left), whereas benzene and the PAH were not metabolized during 100 days of incubation. In the presence of iron(III), stimulated biodegradation of the model compounds was observed (Fig. 5.3-53, right). Toluene and ethylbenzene were transformed with faster kinetics than under sulfate-reducing conditions. Benzene and naphthalene were also degraded. Of the heterocycles, only benzofuran was degraded under sulfate-reducing conditions. Iron-reducing conditions led to the degradation of benzothiophene and dibenzothiophene, whereas the latter was only degraded when additional nutrients were available. Under nitrate-reducing conditions degradation of dibenzofuran was also observed (Table 5.3-36). The fastest degradation was observed in the presence of oxygen. After the addition of air, all BTEX, PAH and heterocycles were degraded to concentrations below the detection limit within 14 days (Table 5.3-36).
898
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
Fig. 5.3-53: Stürmlinger sand pit site, degradation of benzene, toluene, ethylbenzene and polycyclic aromatic hydrocarbons (PAH) under sulfate-reducing (left) and iron-reducing (right) conditions Table 5.3-36: Stürmlinger sand pit site, degradation of benzene, toluene, ethylbenzene, PAH (polycyclic aromatic hydrocarbons) and heterocycles in groundwater microcosms under different redox conditions: - = no degradation; (+) = slight degradation; + = degradation of > 50 % of the sterile control Distance to the pit 20 m 110 m 110 m Depth below water table 10 m 10 m 2m predominant redox sulfate-reducing iron-reducing nitrate-reducing/aerobic conditions in the field redox conditions in the sulfate-reducing iron-reducing nitrate-reducing aerobic microcosms 1 addition of Fe(II)/Fe(III) oxygen degradation of benzene +2 + toluene + + + + ethylbenzene (+) + + + naphthalene + +2 +2 + acenaphthene +2 + 2 phenanthrene + + pyrene + benzofuran + n.i. n.i. n.i. dibenzofuran +2 + benzothiophene +2,3 (+)2 + dibenzothiophene +2 +2,3 + carbazole + 1 Fe(III) was available for the first 200 days, later sulfate was consumed. 2 Only degraded in microcosms with additional nutrient (ammonia, phosphate, microelements). 3 Slight degradation was observed without nutrients. n.i. = not investigated
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
899
The microcosm experiments demonstrated that anaerobic biodegradation contributes significantly to contaminant elimination in the field. However, the efficiency of the biodegradation varied, depending on which electron acceptors were available (Table 5.3-36). It is clear that sulfate-reducing and especially iron-reducing microbial communities contribute significantly to the elimination of aromatic hydrocarbons in the anaerobic zones of the plume and can even degrade benzene, a persistent priority pollutant. All model contaminants were degraded rapidly in the presence of oxygen. This indicates complete biodegradation at the plume fringes and adjacent to the groundwater fluctuation zone, where nitrate and oxygen are available. Investigation of the Stürmlinger sand pit site showed that in-situ biodegradation in the plume originating from the landfill for municipal waste and gasworks residues resulted in the formation of different redox zones. Three-dimensional hydrogeochemical analysis and pollutant profiles indicated that both anaerobic and aerobic processes contribute to the degradation of BTEX, PAH and heterocycles. However, the reduction of aromatic hydrocarbons could not be related to specific redox processes on the basis of the field data alone. Corresponding microcosm experiments demonstrated that toluene, ethylbenzene, naphthalene, and benzofuran were degradable under sulfate-reducing conditions. Benzene, phenanthrene, benzothiophene, and dibenzothiophene were degraded in the presence of iron(III). In the presence of oxygen, the anaerobically resistant contaminants were also degraded. Due to low retardation and slow biodegradation, acenaphthene, dibenzofuran, and dibenzothiophene were suitable compounds for monitoring the maximum plume length at this site. Chloroethene polluted site Site description: The Killisfeld site (Fig. 5.3-54) is in an industrial area in Karlsruhe, Germany, in the Rhine river valley (MÜLLER et al., 2006). The highly permeable sand and gravel aquifer in the area (hydraulic conductivity kf = 1.5 - 3.5 10-3 m s-1; groundwater velocity: 0.4 - 0.9 m d-1; water table: 2 - 3 m below surface; aquifer thickness: 12 - 16 m) is interspersed with less permeable areas rich in humic acids. There is a chloroethene plume in the aquifer originating from at least three different source areas. There are also several domestic waste landfills downgradient from the chloroethene source areas from which organic substances infiltrate into the groundwater.
900
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
Fig. 5.3-54: Killisfeld site, chloroethene contamination (average vertical concentrations) originating from three source zones (S1-S3), locations of domestic waste disposal sites and investigated wells (with permission of G.U.C. GEO UMWELT CONSULT GmbH, Karlsruhe, Germany)
Contaminant profiles: The groundwater is contaminated with the chlorinated ethenes perchloroethene (PCE) and trichloroethene (TCE) from at least three source zones (S1-S3 in Fig. 5.3-54), resulting in three parallel plumes close to the sources. Further downgradient the three plumes merge and in total reach a plume length of at least 2.5 km. Immediately downgradient from the source zones, chloroethene concentrations of > 2000 µg L-1 have been determined. For approximately the first 600 m of the flow path the contamination mainly consists of PCE and to a lesser extent TCE and cis-1,2-dichloroethene (cDCE). Vinyl chloride (VC) is not detected until the plumes pass the disposal sites further downgradient, where it is the predominating contaminant. Ethene and ethane, dechlorinated end-products of microbial reductive degradation, are also detected downgradient from the disposal sites (see Fig. 5.3-55 for the northern part of the plume). Since the chloroethene contamination decreases along the flow path and the typical microbial degradation products cDCE, VC, ethene, and ethane are observed downgradient from the disposal sites, microbial degradation processes can be assumed to occur at this site.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
901
Fig. 5.3-55: Killisfeld site, distribution of chloroethenes, ethene and ethane (average vertical concentrations) along the flow path of the northern plume
Redox zonation: A distinct redox zonation is observed in all three plumes downgradient from the source zones (see Fig. 5.3-56). Nitrate and nitrite are detected close to the source zones along with a relatively high redox potential of 288 - 343 mV, indicating nitrate-reducing conditions. Iron-/manganesereducing conditions can be deduced from the detection of increased concentrations of the reduced, soluble iron and manganese forms iron(II) and manganese(II) 700 - 1000 m from the source. The soluble iron and manganese ions are formed by reduction of iron(III) and manganese(IV) minerals in the aquifer. Dissolved sulfide was not detected in the groundwater samples. However, iron(II) sulfide was found in the sediment. Since iron(III) as well as sulfate-reducing microorganisms were detected in the plume (Fig. 5.3-57), both processes are most probably occurring in a mixed iron(III)-/sulfatereducing area. Organic substances (elevated total organic carbon) from the disposal sites create locally more strongly reducing methanogenic conditions (detection of up to 2000 µg L-1 methane). The strongly reducing regions of the aquifer are favorable for the reductive dechlorination of the chloroethenes. Therefore, VC is mainly detected in methanogenic zones. The prevailing redox conditions indicate that aerobic degradation is possible only at the plume fringes, i.e. further downgradient and in the groundwater fluctuation zones, where oxygen is available.
902
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
Fig. 5.3-56: Killisfeld site, schematic representation of the redox zones along the northern plume, aerobic zone: O2 t 0.5 mg L-1; nitrate-reducing zone: NO3– t 1 mg L-1; iron/manganese-/sulfate-reducing zone: Fe(II) t 1 mg L-1, Mn(II) > 0.2 mg L-1; methanogenic zone: CH4 t 0.5 mg L-1
Microbial numbers: A microbial survey was conducted using several mostprobable-number (MPN) tests on sediment samples taken at different depths during construction of four groundwater wells (see Fig. 5.3-54 for well locations): - 9071-K4: downgradient close to the chloroethene source; moderately reducing conditions; detection of PCE, TCE and cDCE. - 9071-K5: downgradient close to one of the disposal sites; strongly reducing conditions; detection of VC and ethane. - 8970-K2: further downgradient; reducing conditions; cDCE and VC predominate. - 8970-K1: further downgradient; reducing conditions; cDCE and VC predominate. The microbial numbers of total aerobic heterotrophs, BTEX (benzene, toluene, ethylbenzene, o-, m-, p-xylene) degraders, ammonia- and nitrite-oxidizing bacteria, and methanotrophic bacteria under aerobic conditions were determined. Anaerobic nitrate-/iron(III)-/sulfate-reducing bacteria and methanogenic bacteria were investigated (see Fig. 5.3-57 for results of two wells).
Fig. 5.3-57: Killisfeld site, aerobic and anaerobic microbial numbers (per g of sediment dry substance (DS)) of two wells at three different depths (n.d. = not detected, BTEX = benzene, toluene, ethylbenzene, o-, m-, p-xylene)
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
903
904
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
All four wells investigated showed very high microbial numbers in regions where silt/clay/peat sediments predominate. High numbers of aerobic, BTEXdegrading, nitrifying and methanotrophic bacteria were detected. Microbial numbers decreased with increasing depth. The microbial community composition also shows differences depending on well location within the plume. In well 9071-K5, located in a strongly reducing region of the aquifer downgradient from the disposal sites, methanogenic but no nitrifying bacteria and only a few methanotrophic bacteria were detected. Large numbers of aerobic organisms were detected further downgradient in well 8970-K2, where less reducing conditions occurred. Degradation processes: Aerobic and anaerobic dechlorination was observed with groundwater from different parts of the plume. Results are shown for the following two wells: - 9071-36: downgradient close to a disposal site; strongly reducing conditions; detection of VC and ethane. - 8971-13: further downgradient; reducing conditions; VC predominates. Perchloroethene (9 µmol L-1) was completely dechlorinated during anaerobic incubation with acetate and hydrogen as electron donors (Fig. 5.3-58). The intermediary metabolites TCE and cDCE were detected, but not VC and ethene. It is assumed that VC and ethene were formed in concentrations below the detection limit (306 µg L-1 = 4.9 µmol L-1 for VC and 133 µg L-1 = 4.7 µmol L-1 for ethene) due to the relatively low initial concentration of PCE. In other degradation studies with repeated addition of PCE (cumulative addition of approximately 40 µmol L-1), formation of VC and ethene was demonstrated (Fig. 5.3-59).
Fig. 5.3-58: Killisfeld site, reductive dechlorination in groundwater from different zones of the plume (PCE = perchloroethene, TCE = trichloroethene, cDCE = cis-1,2-dichloroethene, VC = vinyl chloride)
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
905
Groundwater from well 9071-36 directly downgradient from a disposal site in a strongly reducing zone of the plume showed faster dechlorination than groundwater from well 8971-13 further downgradient. A lag phase of about 12 weeks occurred with groundwater from well 8971-13 before the onset of dechlorination (Fig. 5.3-58). Addition of yeast extract, trace elements and vitamins did not result in an increase in the dechlorination rate (data not shown). Hence, groundwater from the investigated wells contained all nutrients required by dechlorinating microorganisms. The lower chlorinated metabolites of reductive dechlorination were also degradable under aerobic conditions (Fig. 5.3-60). Again, degradation of cDCE (4.1 µmol L-1) and VC (19.2 µmol L-1) was faster with groundwater from well 9071-36 close to the source zones. The increasing degradation rate of VC after repeated spiking with VC indicates productive degradation of this compound. Cis-1,2-dichloroethene was degraded as long as VC was available in the microcosm, suggesting co-metabolic degradation of cDCE with VC as auxiliary substrate.
Fig. 5.3-59: Killisfeld site, formation of vinyl chloride (VC) and ethene during reductive dechlorination of perchloroethene (PCE) (TCE = trichloroethene, cDCE = cis-1,2dichloroethene)
906
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
Fig. 5.3-60: Killisfeld site, aerobic dechlorination of cis-1,2-dichloroethene (cDCE) and vinyl chloride (VC) in groundwater from different zones of the plume
The results are consistent with site data. The groundwater downgradient from the disposal sites contains VC, ethene and ethane – products of microbial reductive dechlorination. Organic substances from the disposal sites evidently create strongly reducing conditions, and provide auxiliary substrates required for reductive dehalogenation. Aerobic degradation is possible at the plume fringes, i.e. further downgradient and in the groundwater fluctuation zones, where oxygen is available. Analysis of specific degrading bacteria: Groundwater samples were assayed for the halorespiring microorganisms Dehalobacter sp., Desulfuromonas sp., Desulfomonile tiedjei, Desulfitobacterium sp., and Dehalococcoides sp. (BRADLEY, 2003) with nested polymerase chain reaction (PCR). All sampling wells are in the plume downgradient from the source areas. DNA sequences from Desulfuromonas sp., Desulfomonile tiedjei and Dehalococcoides sp. were detected in groundwater samples (Fig. 5.3-61). Bacteria belonging to the Dehalococcoides cluster are the only microorganisms reported so far to catalyze the complete reductive dechlorination of PCE to ethene (MAYMÓ-GATELL et al., 1997; FETZNER, 1998; MIDDELDORP et al., 1999; BRADLEY, 2003). Their presence is, therefore, of particular interest for sites contaminated with chloroethenes, since they are often detected at sites where complete reductive dechlorination occurred (HENDRICKSON et al., 2002; SCHMIDT et al., 2006a). At the Killisfeld site Dehalococcoides were even found in areas within the plume where no VC or ethene were detected and, therefore, the contaminant distribution did not indicate the presence of Dehalococcoides. There are two possible explanations for this: (i) Since 16S-PCR detection of Dehalococcoides does not differentiate between the various strains, it is possible that incompletely dechlorinating strains were also detected (LÖFFLER & RITALAHTI, 2005) and (ii) another explanation might be that the detected Dehalococcoides sequences reflect previous site conditions more favorable for the growth of Dehalococcoides (HENDRICKSON et al., 2002).
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
907
Fig. 5.3-61: Killisfeld site, polymerase-chain-reaction (PCR) detection of halorespiring bacteria in groundwater
Moreover, PCR allowed growth of Dehalococcoides during dechlorination of PCE to be monitored in a microcosm study. The microcosm was assayed for concentrations of chloroethenes and auxiliary substrates, total cell numbers (TCN), protein content and specific Dehalococcoides sequences (Fig. 5.3-62) at different points in time during degradation. Figure 5.3-62A shows the complete degradation of PCE by reductive dechlorination consuming the electron donors acetate and hydrogen (Fig. 5.3-62B). VC and ethene were detected in only trace amounts below the limit of quantification and are, therefore, not shown on the graph. Growth of microorganisms is demonstrated by increasing total cell number (TCN) and protein content (Fig. 5.3-62C). The increase in the amount of the Dehalococcoides-specific 16S-PCR product on day 16 of incubation (Fig. 5.3-62D) demonstrates that the number of Dehalococcoides increased during PCE degradation.
908
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
Fig. 5.3-62: Killisfeld site, degradation of perchloroethene (PCE) by a Dehalococcoides containing a mixed enrichment culture, A: chloroethene concentrations; B: acetate and hydrogen concentrations; C: total cell number (TCN) and protein content; D: direct 16S-PCR (polymerase chain reaction) (TCE = trichloroethene, cDCE = cis-1,2dichloroethene)
In order to prove the assumption that Dehalococcoides bacteria only occur when complete reductive dechlorination takes place, incomplete dechlorinating microcosms from another chloroethene contaminated site (Frankenthal, Germany) were compared with Killisfeld microcosms. Six reductively dechlorinating groundwater microcosms from the Killisfeld site
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
909
and nine from the Frankenthal site were analyzed for rDNA sequences of Dehalococcoides with nested 16S-PCR. All microcosms originating from the Killisfeld site showed complete anaerobic degradation of PCE, TCE and cDCE (see Fig. 5.3-63, left). In contrast, the Frankenthal microcosms only dechlorinated PCE via TCE to cDCE, which accumulated (see Fig. 5.3-63, right). The 16S-PCR results correspond to the observed dechlorinating capabilities of the microcosms. Organisms of the Dehalococcoides cluster were detected only in completely dechlorinating microcosms (Fig. 5.3-63, bottom). These results confirm previous reports that only Dehalococcoides are capable of degrading PCE completely to ethene (MAYMÓ-GATELL et al., 1997; FETZNER, 1998; MIDDELDORP et al., 1999; BRADLEY, 2003). Further comparison of the two sites Killisfeld and Frankenthal has shown that contaminant profiles, hydrochemical site characteristics, dechlorination activity in microcosms, and the presence of Dehalococcoides correlated well (Table 5.3-37). Bacteria belonging to the Dehalococcoides cluster were only detected in microcosms where complete reductive dechlorination was observed. Other dechlorinating microorganisms were detected at both sites. 16S-PCR detection of specific contaminant-degrading bacteria may complement analyses of contaminant profiles at contaminated sites and may be used instead of rather long degradation studies in the laboratory.
Fig. 5.3-63: Killisfeld and Frankenthal sites, dechlorination activity and 16S-PCR (polymerase chain reaction) results of groundwater microcosms (PCE = perchloroethene, TCE = trichloroethene, cDCE = cis-1,2-dichloroethene)
910
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
Table 5.3-37: Killisfeld and Frankenthal sites, site characteristics and polymerase chain reaction (PCR) results (PCE = perchloroethene, TCE = trichloroethene, cDCE = cis-1,2dichloroethene, ORP = oxidation-reduction-potential) Killisfeld
Frankenthal
primary contaminants on site PCE and TCE PCE and TCE redox conditions on site (ORP) 47 - 469 mV 140 - 555 mV degradation products detected on site cDCE, VC, ethene mainly cDCE dechlorination in groundwater PCE o o o o ethene PCE o o cDCE microcosms 16S-PCR detection of Dehalococcoides (PCE o ethene) in vertically averaged positive negative1 groundwater samples and microcosms 16S-PCR detection of other halorespirers positive positive (PCE o cDCE) in groundwater samples and microcosms 1 Dehalococcoides were only found in one of 17 investigated wells when multilevel sampling techniques were used. Only in this area was vinyl chloride (VC) also detected.
Isotope fractionation: Selected aerobic microcosms from the Killisfeld site were assayed for isotope fractionation. Significant fractionation was observed during aerobic degradation of cDCE with VC (Fig. 5.3-64). Analyses for the Frankenthal site also showed isotope fractionation during anaerobic degradation of PCE and TCE (data not shown). The enrichment factors can be used for conservative quantification of biodegradation in the field (MARTIN et al., 2006).
Fig. 5.3-64: Killisfeld site, isotope fractionation during aerobic degradation of cis-1,2dichloroethene (cDCE) and vinyl chloride (VC)
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
911
At the Killisfeld site, the analysis of the hydrochemical data showed the occurrence of strongly reducing conditions along with elevated total organic carbon (TOC) (possible auxiliary substances), indicating site conditions favorable for reductive dechlorination of the contaminating chloroethenes. The detection of specific metabolites of microbial degradation (cDCE, VC and ethene) then demonstrated that microbial dechlorination occurs at this site. Investigation of microbial numbers and detection of specific halorespiring microorganisms further confirmed that microorganisms with the metabolic traits needed for degradation are present and viable in the field. Finally, microcosm studies proved that the autochthonous microflora is able to reductively degrade PCE completely to ethene under anaerobic conditions and to degrade cDCE and VC under aerobic conditions. The applied methods complemented each other. The results led to the conclusion that reductive dechlorination of PCE to TCE, cDCE, VC, ethene, and ethane occurred in the strongly anaerobic parts of the plume and that the lower chlorinated metabolites are partly degraded in regions where oxygen is available (plume fringes). Summary of case studies: The results obtained at the Stürmlinger sand pit and Killisfeld sites demonstrate that the different assessment methods complement each other and allow a well-founded evaluation of the degradation processes occurring in the field. The different techniques can be combined depending on the specific site conditions. Similar strategies are recommended for other classes of biodegradable contaminants. References and further reading ACTON, D. W. & BARKER, J. F. (1992): In-situ biodegradation potential of aromatic hydrocarbons in anaerobic groundwaters. J. Contamin. Hydrol., 9, 325-352. ADAMS, J. M. & EVANS, S. (1979): Determination of the cation-exchange capacity (layer charge) of small quantities of clay minerals by nephelometry. Clays Clay Minerals, 27, 137-139. AG BODEN (1996): Bodenkundliche Kartieranleitung, 4. Aufl. (AG Boden (“Soil” Working Group of the German Federal and State Geological Surveys) (1996). Soil mapping manual, 4th edn.). ALLABY, M. (1977): A Dictionary of the Environment. Van Nostrand Reinhold, New York. ALTES, J. (1976): Die Grenztiefe bei Setzungsberechnungen. Bauingenieur, 51, 93-96. ALTFELDER S., STRECK T. & RICHTER J. J. (1999): Effect of air-drying on sorption kinetics of the herbicide Chlorotoluron in soil. J. Environ. Qual., 28, 1154-1161. ALTFELDER, S. & STRECK, T. (2006): Capability and limitations of first-order and diffusion approaches to describe long-term sorption of chlortoluron in soil. J. Contam. Hydrol., 86, 279-298.
912
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
ALTHOFF, K., MUNDT, M., EISENTRAEGER, A., DOTT, W. & HOLLENDER, J. (2001): Microcosm-experiments to assess the potential for natural attenuation of contaminated groundwater. Water Res., 35, 3, 720-728. AMONETTE, J. E. & ZELAZNY, L. W. (Eds.) (1994): Quantitative Methods in Soil Mineralogy. Miscellaneous Publication, Soil Science Society of America, Madison. ANDREWS, J. E., BRIMBLECOMBE, P., JICKELLS, T., LISS, P. & REID, B. (2004): An Introduction to Environmental Chemistry. Blackwell. APPELO, C.A., VAN DER WEIDEN, M. J., TOURNASSAT, C. & CHARLET, L. (2002): Surface complexation of ferrous iron and carbonate on ferrihydrite and the mobilization of arsenic. Environ. Sci. Technol., 36, 14, 3096-3103. APPELO, C. A. J.; POSTMA, D. (2005): Geochemistry, groundwater and pollution. 2nd edn., A. A. Balkema Publishers AULENTA, F., Bianchi, A., Majone, M., Petrangeli Papini, M., Potalivo M. & Tandoi V. (2005): Assessment of natural or enhanced in-situ bioremediation at a chlorinated solvent-contaminated aquifer in Italy: a microcosm study. Environ Int., 31, 2, 185-190. Aulenta, F., MAJONE, M. VERBO, P. & TANDOI, V. (2002): Complete dechlorination of tetrachloroethene to ethene in presence of methanogenesis and acetogenesis by an anaerobic sediment microcosm. Biodegradation, 13, 411-424. BACHE, B. W. (1970): Barium isotope method for measuring cation-exchange capacity of soils and clays. J. Sci. Food and Agric., 21, 169-171. BACHE, B. W. (1976): The measurement of cation exchange capacity of soils. J. Sci. Food Agric., 27, 273-280. BAILEY, S. W. (1980): Summary of recommendations of AIPEA nomenclature committee on clay minerals. American Mineralogist, 65, 1-7. BANNERT, M., BERGER, W., FISCHER, H., HORCHLER, D., KEESE, K., LEHNIK-HABRINK, P., LÜCK, D., PRITZKOW, J. & WIN, T. (2001): Anforderungen an Probennahme Probenvorbehandlung und chemische Untersuchungsmethoden auf Bundesliegenschaften. (Eds.), Bundesanstalt für Materialforschung und -prüfung (BAM) Berlin, Amts- und Mitteilungsblatt der BAM, Sonderheft 2/2001. www.bam.de/de/service/publikationen/publikationen_medien/probennahme.pdf BARBER, C. & BRIEGEL, D. (1987): A method for the in-situ determination of dissolved methane in groundwater in shallow aquifers. J. of Contaminant Hydrology, 2, 51- 60. BARTH, J. A. C., SLATER, G., SCHÜTH, C., BILL, M., DOWNEY, A., LARKIN, M. & KALIN, R. M. (2002): Carbon isotope fractionation during aerobic biodegradation of trichloroethene by Burkholderia cepacia G4: a tool to map degradation mechanisms. Appl. Environ. Microbiol., 68, 4, 1728-1734. BASCOMB, C. L. (1964): Rapid method for the determination of the cation exchange capacity of calcareous and non-calcareous soils. J. Sci. Food and Agric., 15, 821-823. BEAR, J. (1972): Dynamics of fluids in porous media. Elsevier, New York.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
913
BEHRENS, W. & FEISER, J. (1995): Anmerkungen zur Berechnung der Setzungen von Deponiebauwerken. AbfallwirtschaftsJournal, 7, 9, 545-549. BELLER, H. R. (2002): Analysis of benzylsuccinates in groundwater by liquid chromatography/tandem mass spectrometry and its use for monitoring in situ BTEX biodegradation. Environ. Sci. Technol., 36, 2724-2728. BELLER, H. R., DING, W.-H. & REINHARD, M. (1995): Byproducts of anaerobic alkylbenzene metabolism useful as indicators of in situ bioremediation. Environ. Sci. Technol., 29, 2864-2870. BERGAYA, F., THENG, B. K. G. & LAGALY, G. (Eds.) (2006): Handbook of Clay Science. Elsevier. BERGAYA, F. & VAYER, M. (1997): CEC of clays; measurement by adsorption of a copper ethylenediamine complex. Appl. Clay Sci., 12, 275-280. BIELEFELDT A. R. & STENSEL H. D. (1999): Biodegradation of aromatic compounds and TCE by a filamentous bacteria-dominated consortium. Biodegradation, 10, 1-13. BILITEWSKI, B., HÄRDTLE, G. & MAREK, K. (2000): Abfallwirtschaft (Waste management). 3rd edn. Springer, Berlin. BILITEWSKI, B. (2001): Praktikumsskript „Altlasten“ (Laboratory notes for lecture on hazardous sites). Dresden-Pirna. BJERG, P. L., BRUN, A., NIELSEN, P. H. & CHRISTENSEN, T. H. (1996): Application of a model accounting for kinetic sorption and degradation to in situ microcosm observations on the fate of aromatic hydrocarbons in an anaerobic aquifer. Water Resources Research, 32, 6, 1831-1841. BJERG, P. L., RÜGGE, K., CORTSEN, J., NIELSEN, P. H. & CHRISTENSEN, T. H. (1999): Degradation of aromatic and chlorinated aliphatic hydrocarbons in the anaerobic part of the Grindsted landfill leachate plume: In situ microcosm and laboratory batch experiments. Ground Water, 37, 1, 113-121. BLOOM Y., ARAVENA R., HUNKELER D., EDWARDS E. & FRAPE S. K. (2000): Carbon isotope fractionation during microbial degradation of trichloroethene, cis-1,2dichloroethene, and vinyl chloride: Implications for assessment of natural attenuation. Environ. Sci. Technol., 34, 13, 2768-2772. BORDEN R. C., DANIEL R. A., LEBRUN IV L. E. & DAVIS C. W. (1997): Intrinsic biodegradation of MTBE and BTEX in a gasoline-contaminated aquifer. Water Resources Research, 33, 5, 1105-1115. BORDEN R. C., GOMEZ C. A. & BECKER M. T. (1995): Geochemical indicators of intrinsic bioremediation. Ground Water, 33, 2, 180-189. BOWEN, H. J. M. (1979): Environmental chemistry of the elements. Academy Press, London. BRADLEY, P. M. (2003): History and ecology of chloroethene biodegradation: A review. Bioremediation Journal, 7, 2, 81-109.
914
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
BRAR, S. K. & GUPTA, S. K. (2000): Biodegradation of trichloroethylene in a rotating biological contactor. Water Res., 34, 17, 4207-4214. BREEN, C. & ROCK, B. J. (1991): The competitive adsorption of dyes on clays. 7th Euroclay Conference, Dresden. BRUNAUER, S., EMMETT, P. H. & TELLER, E. (1938): Adsorption of gases in multimolecular layers. J. Amer. Chem. Soc., 60, 309-319. BRUSSEAU, M. L. & RAO P. S. C. (1989): Sorption nonideality during organic contaminant transport in porous media. Crit. Rev. Environ. Control, 19, 33-99. BRUSSEAU, M. L., JESSUP, R. E. & RAO, P. S. C. (1991): Nonequilibrium Sorption of Organic Chemicals: Elucidation of Rate-Limiting Processes. Environ. Sci. and Technol., 25, 134-142. BUHMANN, D. & DREYBRODT, W. (1987): Calcite dissolution kinetics in the system H2OCO2-CaCO3 with participation of foreign ions. Chem. Geology, 64, 89-102. BUJDÁK J. & KOMADEL P. (1997): Interaction of Methylene Blue with reduced charge montmorillonite. J. Phys. Chem. B, 101, 9065. BUWAL (2003): Praxishilfe Grundwasserprobenahme. Bundesamt für Umwelt, Wald und Landschaft, Bern, Switzerland. BWK (2004): Erarbeitung von Leistungsbeschreibungen und Leistungsverzeichnissen zur Grundwasserprobenahme bei Altlasten im Lockergestein. BWK-Merkblatt 5. Gelbdruck. Bund der Ingenieure für Wasserwirtschaft, Abfallwirtschaft und Kulturbau, Pfullingen, Germany. CERNIGLIA, C. E. (1992): Biodegradation of polycyclic aromatic hydrocarbons. Biodegradation, 3, 351-368. CHAPELLE F. H., MCMAHON P. B., DUBROWSKY N. M., FUJII, R. F., OAKSFORD E. T. & VROBLESKY D. A. (1995): Deducing the distribution of terminal electron-accepting processes in hydrologically diverse groundwater systems. Water Resources Research, 31, 2, 359-371. CHAPELLE, F. H. (2001): Ground-water Microbiology and Geochemistry. Wiley & Sons, New York. CHARTRAND, M. M. G., WALLER, A., MATTES, T. E., ELSNER, M., LACRAMPE-COULOUME, G., GOSSETT, J.M., EDWARDS, E. A. & SHERWOOD LOLLAR, B. (2005): Carbon isotopic fractionation during aerobic vinyl chloride degradation. Environ. Sci. Technol., 39, 4, 1064-1070. CHHABRA, R., PLEYSIER, J. & CREMERS, A. (1975): The measurement of the cation exchange capacity and exchangeable cations in soils: A new method. Proc. Int. Clay Conf. 1975, Wilmette, Illinois, 439-449. CHILDS, C. W. (1992): Ferrihydrite: A review of structure, properties and occurrence in relation to soils. Z. Pflanzenernähr. Bodenk., 155, 441-448. CHIOU, C. T., PETERS L. J., & FREED, V. H. (1979): A Physical Concept of Soil-Water Equilibria for Non-Ionic Compounds Science, 206, 831-832.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
915
CHIOU, C. T., PORTER P. E. & SCHMEDDLING D. W. (1983): Partition equilibria of nonionic organic compounds between soil organic matter and water. Environ. Sci. Technol., 17, 227-231. CHO, J. S., WILSON, J. T., DIGIULIO, D. C., VARDY, J. A. & CHOI, W. (1997): Implementation of natural attenuation at a JP-4 jet fuel release after active remediation. Biodegradation, 8, 265-273. CHRISTENSEN, T. H., BJERG, P. L. & KJELDSEN, P. (2001): Natural attenuation as an approach to remediation of groundwater pollution at landfills. In: Treatment of Contaminated Soil, STEGMANN, R., BRUNNER, G., CALMANO, W. & MATZ, G. (Eds.), Springer, Berlin, 587-602. CHU, K.-H., MAHENDRA, S., SONG, D. L., CONRAD, M.E. & ALVAREZ-COHEN, L. (2004): Stable carbon isotope fractionation during aerobic biodegradation of chlorinated ethenes. Environ. Sci. Technol., 38, 11, 3126-3130. COHEN, R. M. & MERCER, J. W. (1993): DNAPL site evaluation. Smoley, Boca Raton, FL. COLEMAN, N. V., MATTES, T. E., GOSSETT, J. M. & SPAIN, J. C. (2002a): Biodegradation of cis-dichloroethene as the sole carbon source by a ȕ-proteobacterium. Appl. Environ. Microbiol., 68, 6, 2726-2730. COLEMAN, N. V., MATTES, T. E., GOSSETT, J. M. & SPAIN, J. C. (2002b): Phylogenetic and kinetic diversity of aerobic vinyl chloride-assimilating bacteria from contaminated sites. Appl. Environ. Microbiol., 68, 12, 6162-6171. CORNELL, R. M. & SCHWERTMANN, U. (2003): The iron oxides: structure, properties, reactions, occurrences and uses. Wiley-VCH, Weinheim. Cotter-Howells, J. D. Campbell, L. S., Valsami-Jones, E. & Batchelder, M. (2000): Environmental Mineralogy: Microbial Interactions, Anthropogenic Influences, Contaminated Land and Waste Management. Mineralogical Society Book Series Vol. 9. COZZARELLI, I. M., BAEDECKER, M. J., EGANHOUSE, R. P. & GOERLITZ, D. F. (1994): The geochemical evolution of low-molecular-weight organic acids derived from the degradation of petroleum contaminants in groundwater. Geochim. Cosmochim. Acta, 58, 2, 863-877. COZZARELLI, I. M., HERMAN, J. S. & BAEDECKER, M. J. (1995): Fate of microbial metabolites of hydrocarbons in a coastal plain aquifer: The role of electron acceptors. Environ. Sci. Technol., 29, 458-469. CREMERS, A. & PLEYSIER, J. (1973a): Adsorption of the silver-thiourea complex in montmorillonite clay. Nature Phys. Sci., 243, 86-87. CREMERS, A. & PLEYSIER, J. (1973b): Coordination of silver in silver-thiourea montmorillonite. Nature Phys. Sci., 244, 93. DANKO, A. S., LUO, M., BAGWELL, C. E., BRIGMON, R. L. & FREEDMAN, D. L. (2004): Involvement of linear plasmids in aerobic biodegradation of vinyl chloride. Appl. Environ. Microbiol., 70, 10, 6092-6097. DAVIS, S. N. & DE WIEST, R. J. M. (1966): Hydrogeology. Wiley & Sons, New York.
916
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
DÉKÀNY, I., SZÁNTÓ, F. & NAGY, L. G. (1978): Selective adsorption of liquid mixtures on organophilic clay minerals. Progress in Colloid and Polymer Science, 65, 125-132. DELLER, B. (1981): Determination of exchangeable acidity, carbonate ions and change of buffer in triethanolamine-buffered solutions percolated through soil samples containing carbonates. Commun. Soil Sci. Plant Analysis, 12, 161-177. DIN 1319-3 Fundamentals of metrology – Part 3: Evaluation of measurements of a single measurand, measurement uncertainty. Beuth, Berlin. DIN 1319-4 Fundamentals of metrology – Part 4: Evaluation of measurements; uncertainty of measurement. Beuth, Berlin. DIN 4019-1, Subsoil; Settlement Calculations for Perpendicular Central Loading. Beuth, Berlin. DIN 4021 Soil; exploration by excavation and borings; sampling. Beuth, Berlin. DIN 4022-1 Subsoil and groundwater; classification and description of soil and rock; borehole logging of soil and rock not involving continuous core sample recovery. Beuth, Berlin. DIN 4022-2 Subsoil and groundwater; Designation and description of soil types and rock; Stratigraphic representation for borings in rock. Beuth, Berlin. DIN 4022-3 Subsoil and groundwater; Designation and description of soil types and rock; Borehole log for boring in soil (loose rock) by continuous extraction of cores. Beuth, Berlin. DIN 4023 Borehole logging; graphical representation of the results. Beuth, Berlin. DIN 4030-1 Assessment of water, soil and gases for their aggressiveness to concrete; principles and limiting values. Beuth, Berlin. DIN 4030-2 Assessment of water, soil and gases for their aggressiveness to concrete; collection and examination of water and soil samples. Beuth, Berlin. DIN 4049-1 Hydrology; basic terms. Beuth, Berlin. DIN 4049-2 Hydrology; terms relating to quality of waters. Beuth, Berlin. DIN 4049-3 Hydrology – Part 3: Terms for the quantitative hydrology. Beuth, Berlin. DIN 4084 Subsoil; Calculations of terrain rupture and slope rupture. Beuth, Berlin. DIN 4124 Excavations and trenches – Slopes, planking and strutting, breadths of working spaces. Beuth, Berlin. DIN 18123 Soil, investigation and testing – Determination of grain-size distribution. Beuth, Berlin. DIN 18129 Soil, investigation and testing – Determination of lime content. Beuth, Berlin. DIN 19682-2 Methods of soil investigations for agricultural water engineering – Field tests – Part 2: Determination of soil texture. Beuth, Berlin.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
917
DIN 19684-3 Methods of soil investigations for agricultural water engineering – Chemical laboratory tests – Part 3: Determination of the loss on ignition and the residue of soil after ignition. Beuth, Berlin. DIN 19711 Graphical and letter symbols for hydrogeology. Beuth, Berlin. DIN 19730 Soil quality – Extraction of trace elements with ammonium nitrate solution. Beuth, Berlin. DIN 19731 Soil quality – Utilization of soil material. Beuth, Berlin. DIN 19734 Soil quality – Determination of chromium(VI) in phosphate extract. Beuth, Berlin. DIN 19738 Soil quality – Absorption availability of organic and inorganic pollutants from contaminated soil material. Beuth, Berlin. DIN 32645 Chemical analysis; decision limit; detection limit and determination limit; estimation in case of repeatability; terms, methods, evaluation. Beuth, Berlin. DIN 38402-12 German standard methods for the examination of water, waste water and sludge; general information (group A); sampling from barrages and lakes (A 12). Beuth, Berlin. DIN 38402-13 German standard methods for the examination of water, waste water and sludge; general information (group A); sampling from aquifers (A 13). Beuth, Berlin. DIN 38402-15 German standard methods for the examination of water, waste water and sludge; general information (group A); sampling of flowing waters (A 15). Beuth, Berlin. DIN 38404-5 German standard methods for examination of water, waste water and sludge; physical and physico-chemical characteristics (group C); determination of pH value (C5). Beuth, Berlin. DIN 38405-1 German standard methods for the examination of water, waste water and sludge; anions (group D); determination of chloride ions (D 1). Beuth, Berlin. DIN 38405-4 German standard methods for the examination of water, waste water and sludge; anions (group D); determination of fluoride (D 4). Beuth, Berlin. DIN 38405-5 German standard methods for the examination of water, waste water and sludge; anions (group D); determination of sulfate ions (D 5). Beuth, Berlin. DIN 38405-9 German standard methods for examination of water, waste water and sludge; anions (group D), determination of nitrate ion (D9). Beuth, Berlin. DIN 38405-13 German Standard Methods for the Analysis of Water, Waste Water and Sludge; Anions (Group D); Determination of Cyanides (D 13). Beuth, Berlin. DIN 38405-14 German standard methods for the examination of water, waste water and sludge; anions (group D); determination of cyanides in drinking water, and in groundwater and surface water with low pollution levels (D 14).
918
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
DIN 38405-23 German standard methods for the examination of water, waste water and sludge – Anions (Group D) – Part 23: Determination of selenium by atomic absorption spectrometry (D 23). Beuth, Berlin. DIN 38405-24 German standard methods for the examination of water, waste water and sludge; anions (group D); photometric determination of chromium(VI) using 1,5-diphenylcarbonohydrazide (D 24). Beuth, Berlin. DIN 38405-26 German standard methods for the examination of water, waste water and sludge; anions (group D); determination of dissolved sulfide by spectrometry (D 26). Beuth, Berlin. DIN 38405-32 German standard methods for the examination of water, waste water and sludge – Anions (group D) – Part 32: Determination of antimony by atomic absorption spectrometry (D 32). Beuth, Berlin. DIN 38406-5 German standard methods for the examination of water, waste water and sludge; cations (group E); determination of ammonia-nitrogen (E 5). Beuth, Berlin. DIN 38406-6 German standard methods for the examination of water, waste water and sludge – Cations (group E) – Part 6: Determination of lead by atomic absorption spectrometry (AAS) (E 6). Beuth, Berlin. DIN 38406-7 German standard methods for the examination of water, waste water and sludge; cations (group E); determination of copper by atomic absorption spectrometry (AAS) (E 7). Beuth, Berlin. DIN 38406-E7 German standard methods for the examination of water, waste water and sludge; cations (group E); determination of copper by atomic absorption spectrometry (AAS) (E 7). Beuth, Berlin. DIN 38406-8 German standard methods for examination of water, waste water and sludge – Cations (group E) – Part 8: Determination of zinc – Method by atomic absorption spectrometry (AAS) using an air-ethine flame (E 8). Beuth, Berlin. DIN 38406-11 German standard methods for the examination of water, waste water and sludge; cations (group E); determination of nickel by atomic absorption spectrometry (AAS) (E 11). Beuth, Berlin. DIN 38406-13 German standard methods for the examination of water, waste water and sludge; cations (group E); determination of potassium by atomic absorption spectrometry (AAS) using an air-acetylene flame (E 13). Beuth, Berlin. DIN 38406-14 German standard methods for the examination of water, waste water and sludge; cations (group E); determination of sodium by atomic absorption spectrometry (ASS) using an air-acetylene flame (E 14). Beuth, Berlin. DIN 38406-24 German standard methods for the examination of water, waste water and sludge; cations (group E); determination of cobalt by atomic absorption spectrometry (AAS) (E 24). Beuth, Berlin. DIN 38406-26 German standard methods for the examination of water, waste water and sludge – Cations (group E) – Part 26: Determination of thallium by atomic absorption spectrometry (AAS) using electrothermal atomisation (E 26). Beuth, Berlin.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
919
DIN 38406-28 German standard methods for the examination of water, waste water and sludge – Cations (group E) – Part 28: Determination of dissolved barium by atomic absorption spectrometry (E 28). Beuth, Berlin. DIN 38406-32 German standard methods for the examination of water, waste water and sludge – Cations (group E) – Part 32: Determination of iron by atomic absorption spectrometry (E 32). Beuth, Berlin. DIN 38406-33 German standard methods for the examination of water, waste water and sludge – Cations (group E) – Part 33: Determination of manganese by atomic absorption spectrometry (E 33). Beuth, Berlin. DIN 38407-2 German standard methods for the determination of water, waste water and sludge; jointly determinable substances (group F); determination of low volatile halogenated hydrocarbons by gas chromatography (F 2). Beuth, Berlin. DIN 38407-3 German standard methods for the determination of water, waste water and sludge – Jointly determinable substances (group F) – Part 3: Determination of polychlorinated biphenyls (F 3). Beuth, Berlin. DIN 38407-8 German standard methods for the examination of water, waste water and sludge – Jointly determinable substances (group F) – Part 8: Determination of 6 polynuclear aromatic hydrocarbons (PAH) in water by high performance liquid chromatography (HPLC) with fluorescence detection (F 8). Beuth, Berlin. DIN 38407-9 German standard methods for the examination of water, waste water and sludge; substance group analysis (group F); determination of benzene and some of its derivatives by gas chromatography (F 9). Beuth, Berlin. DIN 38407-14 German standard methods for the examination of water, waste water, and sludge – Jointly determinable substances (group F) – Part 14: Determination of phenoxyalkyl carbonic acids by gas chromatography and mass-spectrometric detection after solid-liquid-extraction and derivatization (F 14). Beuth, Berlin. DIN 38409-1 German standard methods for the examination of water, waste water and sludge; parameters characterizing effects and substances (group H); determination of total dry residue, filtrate dry residue and residue on ignition (H 1). Beuth, Berlin. DIN 38409-8 German standard methods for the examination of water, waste water and sludge; summary indices of actions and substances (group H); determination of extractable organically bonded halogens (EOX) (H 8). Beuth, Berlin. DIN 38409-16 German standard methods for the examination of water, waste water and sludge; general measures of effects and substances (group H); determination of the phenol index (H 16). Beuth, Berlin. DIN 38409-41 German Standard Methods for Examination of Water, Waste Water and Sludge; Summary Action and Material Characteristic Parameters (Group H); Determination of the Chemical Oxygen Demand (COD) in the Range over 15 mg/l (H41). Beuth, Berlin. DIN 38413-2 German standard methods for the examination of water, waste water and sludge; individual constituents (group P); determination of vinyl chloride by headspace gas chromatography (P 2). Beuth, Berlin.
920
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
DIN 38414-4 German standard methods for the examination of water, waste water and sludge; sludge and sediments (group S); determination of leachability by water (S 4). Beuth, Berlin. DIN 38414-17 German standard methods for the examination of water, waste water and sludge; sludge and sediments (group S); determination of strippable and extractable organically bound halogens (S 17). Beuth, Berlin. DIN 38414-18 German standard methods for the examination of water, waste water and sludge; sludge and sediments (group S); determination of adsorbed organically bound halogens (AOX) (S 18). Beuth, Berlin. DIN 38414-20 German standard methods for the examination of water, waste water and sludge – Sludge and sediments (group S) – Part 20: Determination of 6 polychlorinated biphenyls (PCB) (S 20). Beuth, Berlin. DIN 38414-22 German standard methods for the examination of water, waste water and sludge – Sludge and sediments (group S) – Part 22: Determination of dry residue by freezing and preparation of the freeze dried mass of sludge (S 22). Beuth, Berlin. DIN 38414-23 German standard methods for the examination of water, waste water and sludge – Sludge and sediments (group S) – Part 23: Determination of 15 polycyclic aromatic hydrocarbons (PAH) by high performance liquid chromatography (HPLC) and fluorescence detection (S 23). Beuth, Berlin. DIN 38414-24 German standard methods for the examination of water, waste water and sludge – Sludge and sediments (group S) – Part 24: Determination of polychlorinated dibenzodioxins (PCDD) and polychlorinated dibenzofuranes (PCDF) (S 24). Beuth, Berlin. DIN 51084 Testing of oxidic raw materials for ceramic, glass and glazes; determination of fluoride content. Beuth, Berlin. DIN 51527-1 Testing of petroleum products; determination of polychlorinated biphenyls (PCB); preseparation by liquid chromatography and determination of six selected PCB compounds by gas chromatography using an electron capture detector DIN EN 932-1 Test for general properties of aggregates – Part 1: Methods for sampling; German version EN 932-1:1996. Beuth, Berlin. DIN EN 1233 Water quality – Determination of chromium – Atomic absorption spectrometric methods. Beuth, Berlin. DIN EN 1483 Water quality – Determination of mercury. Beuth, Berlin. DIN EN 1484 (H14) Water analysis – Guidelines for the determination of total organic carbon (TOC) and dissolved organic carbon (DOC). Beuth, Berlin. DIN EN 12338 Water quality – Determination of mercury – Methods after enrichment by amalgamation. Beuth, Berlin. DIN EN 12673 Water quality – Gas chromatographic determination of some selected chlorophenols in water. Beuth, Berlin.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
921
DIN EN 25663 Water quality; determination of Kjeldahl nitrogen; method after mineralization with selenium. Beuth, Berlin. DIN EN 25667-2 Water quality; sampling; part 2: guidance on sampling techniques. Beuth, Berlin. DIN EN 25813 Water quality; determination of dissolved oxygen; iodometric method. Beuth, Berlin. DIN EN 25814 Water quality; determination of dissolved oxygen by the electrochemical probe method. Beuth, Berlin. DIN EN 26777 Water quality; determination of nitrite; molecular absorption spectrometric method. Beuth, Berlin. DIN EN 27888 Water quality; determination of electrical conductivity. Beuth, Berlin. DIN EN ISO 5667-3 Water quality – Sampling – Part 3: Guidance on the preservation and handling of water samples (ISO 5667-3); German version EN ISO 5667-3. Beuth, Berlin. DIN EN ISO 5961 Water quality – Determination of cadmium by atomic absorption spectrometry. Beuth, Berlin. DIN EN ISO 6468 Water quality – Determination of certain organochlorine insecticides, polychlorinated biphenyls and chlorobenzenes – Gas-chromatographic method after liquid-liquid extraction. Beuth, Berlin. DIN EN ISO 7027 Water quality – Determination of turbidity. Beuth, Berlin. DIN EN ISO 7887 Water quality – Examination and determination of colour. Beuth, Berlin. DIN EN ISO 7980 Water quality – Determination of calcium and magnesium – Atomic absorption spectrometric method. Beuth, Berlin. DIN EN ISO 9377-2 Water quality – Determination of hydrocarbon oil index – Part 2: Method using solvent extraction and gas chromatography. Beuth, Berlin. DIN EN ISO 10301 Water quality – Determination of highly volatile halogenated hydrocarbons – Gas-chromatographic methods. Beuth, Berlin. DIN EN ISO 10304-1 Water quality – Determination of dissolved fluoride, chloride, nitrite, orthophosphate, bromide, nitrate and sulfate ions, using liquid chromatography of ions – Part 1: Method for water with low contamination. Beuth, Berlin. DIN EN ISO 10304-3 Water quality – Determination of dissolved anions by liquid chromatography of ions – Part 3: Determination of chromate, iodide, sulfite, thiocyanate and thiosulfate. Beuth, Berlin. DIN EN ISO 11732 Water quality – Determination of ammonium nitrogen – Method by flow analysis (CFA and FIA) and spectrometric detection. Beuth, Berlin. DIN EN ISO 11885 Water quality – Determination of 33 elements by inductively coupled plasma atomic emission spectroscopy. Beuth, Berlin. DIN EN ISO 11969 Water quality – Determination of arsenic – Atomic absorption spectrometric method (hydride technique). Beuth, Berlin.
922
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
DIN EN ISO 10695 Water quality – Determination of selected organic nitrogen and phosphorus compounds – Gas chromatographic methods. Beuth, Berlin. DIN EN ISO 13395 Water quality – Determination of nitrite nitrogen and nitrate nitrogen and the sum of both by flow analysis (CFA and FIA) and spectrometric detection. Beuth, Berlin. DIN EN ISO 14403 Water quality – Determination of total cyanide and free cyanide by continuous flow analysis. Beuth, Berlin. DIN EN ISO 15913 Water quality – Determination of selected phenoxyalkanoic herbicides, including bentazones and hydroxybenzonitriles by gas chromatography and mass spectrometry after solid phase extraction and derivatization. Beuth, Berlin. DIN EN ISO 17294-2 Water quality – Application of inductively coupled plasma mass spectrometry (ICP-MS) – Part 2: Determination of 62 elements (ISO 17294-2:2003); German version EN ISO 17294-2:2004. Beuth, Berlin. DIN ISO 9964-3 Water quality – Determination of sodium and potassium – Part 3: Determination of sodium and potassium by flame emission spectrometry. Beuth, Berlin. DIN ISO 10381-1 Soil quality – Sampling – Part 1: Guidance on the design of sampling programmes. Beuth, Berlin. DIN ISO 10381-2 Soil quality – Sampling – Part 2: Guidance on sampling techniques. Beuth, Berlin. DIN ISO 10381-3 Soil quality – Sampling – Part 3: Guidance on safety. Beuth, Berlin. DIN ISO 10381-4 Soil quality – Sampling – Part 4: Guidance on the procedure for investigation of natural, near-natural and cultivated sites. Beuth, Berlin. DIN ISO 10381-5 Soil quality – Sampling – Part 5: Guidance on the procedure for the investigation of urban and industrial sites with regard to soil contamination. Beuth, Berlin. DIN ISO 10382 Soil quality – Determination of organochlorine pesticides and polychlorinated biphenyls – Gas-chromatographic method with electron capture detection. Beuth, Berlin. DIN ISO 10694 Soil quality – Determination of organic and total carbon after dry combustion (elementary analysis). Beuth, Berlin. DIN ISO 11047 Soil quality – Determination of cadmium, chromium, cobalt, copper, lead, manganese, nickel and zinc in aqua regia extracts of soil – Flame and electrothermal atomic absorption spectrometric methods. Beuth, Berlin. DIN ISO 11074-1 Soil quality – Vocabulary – Part 1: Terms and definitions relating to the protection and pollution of the soil. Beuth, Berlin. DIN ISO 11261 Soil quality – Determination of total nitrogen – Modified Kjeldahl method. Beuth, Berlin. DIN ISO 11262 Soil quality – Determination of cyanide. Beuth, Berlin.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
923
DIN ISO 11264 Soil quality – Determination of herbicides using high performance liquid chromatography with UV-detection. Beuth, Berlin. DIN ISO 11265 Soil quality – Determination of the specific electrical conductivity. Beuth, Berlin. DIN ISO 11272 Soil quality – Determination of dry bulk density. Beuth, Berlin. DIN ISO 11277 Soil quality – Determination of particle size distribution in mineral soil material – Method by sieving and sedimentation. Beuth, Berlin. DIN ISO 11464 Soil quality – Pretreatment of samples for physico-chemical analysis. Beuth, Berlin. DIN ISO 11465 Soil quality – Determination of dry matter and water content on a mass basis – Gravimetric method. Beuth, Berlin. DIN ISO 11466 Soil quality – Extraction of trace elements soluble in aqua regia. Beuth, Berlin. DIN ISO 13877 Soil quality – Determination of polynuclear aromatic hydrocarbons – Method using high-performance liquid chromatographic. Beuth, Berlin. DIN ISO 14154 Soil quality – Determination of some selected chlorophenols in soils – Gaschromatographic method. Beuth, Berlin. DIN ISO 14507 Soil quality – Pretreatment of samples for determination of organic contaminants. Beuth, Berlin. DIN V 4019-100 Soil – Analysis of settlement – Part 100: Analysis in accordance with partial safety factor concept. Pre-standard. Beuth, Berlin. DIN V 19736 (1998): Soil quality – Derivation of concentrations of organic pollutants in soil water (Vornorm). Beuth, Berlin. DOHRMANN, R. (1997): Kationenaustauschkapazität von Tonen – Bewertung bisheriger Analysenverfahren und Vorstellung einer neuen und exakten Silber-ThioharnstoffMethode. PhD thesis RWTH Aachen, AGB-Verlag No. 26. DOHRMANN, R. (2006a): Problems in CEC determination of calcareous clayey sediments using the ammonium acetate method. J. Plant Nutr. Soil Sci., 169, 330 – 334. DOHRMANN, R. (2006b): Cation Exchange Capacity Methodology II: proposal for a modified silver-thiourea method. Appl. Clay Sci., 34, 38-46. DOHRMANN, R. (2006c): Cation Exchange Capacity Methodology III: Correct exchangeable calcium determination of calcareous clays using a new silver-thiourea method. Appl. Clay Sci., 34, 47-57. DOHRMANN, R. (2006d): Cation Exchange Capacity Methodology I: An Efficient Model for the Detection of Incorrect Cation Exchange Capacity and Exchangeable Cation Results. Appl. Clay Sci., 34, 31-37. DRÄGER (2001): Dräger-Röhrchen- / CMS-Handbuch (Manual for indicating test tubes / CMS). Dräger Sicherheitstechnik, Lübeck, Germany. DRESCHER, J. (1997): Deponiebau. Ernst, Berlin.
924
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
DVWK (1992): Entnahme und Untersuchungsumfang von Grundwasserproben. DVWKRegeln 128. DVWK (1997): Tiefenorientierte Probenahme aus Grundwassermessstellen. DVWKMerkblätter 245, Bonn. DYREBORG, S., ARVIN, E. & BROHOLM, K. (1997): Biodegradation of NSO compounds under different redox conditions. J. Cont. Hydrol., 25, 177-197. EL FANTROUSSI, S., NAVEAU, H. & AGATHOS, S. N. (1998): Anaerobic dechlorinating bacteria. Biotechnol. Prog., 14, 167-188. EN ISO 5667-3 (2003): Water quality – Sampling – Part 3: Guidance on the preservation and handling of water samples. International Organization for Standardization, Geneva. EPA (2006a): http://oaspub.epa.gov/wqsdatabase/wqsi_epa_criteria.rep_parameter, downloaded: 13/11/2006. EPA (2006b): EPA On-line Tools for Site Assessment Calculation. www.epa.gov/athens/learn2model/part-two/onsite/ard_onsite.htm EU Heater project (2006): Heater Experiment: Rock and bentonite thermo-hydromechanical (THM) processes in the near field of a thermal source for development of deep underground high level radioactive waste repositories. – Final technical publishable report project: FISS-2001-00024, Contract No. FIKW-CT-2001-00132. FANG, Z. (1995): Flow injection atomic absorption spectrometry. Wiley & Sons, New York. FARMER, V. C. (1974): The infrared Spectra of Minerals. Mineralogical Society Monograph 4, London. FATHEPURE, B. Z., ELANGO, V. K., SINGH, H. & BRUNER, M. A. (2005): Bioaugmentation potential of a vinyl chloride-assimilating Mycobacterium sp., isolated from a chloroethene-contaminated aquifer. FEMS Microbiol. Lett., 248, 227-234. FENNELL, D., CARROl, A., GOSSETT, J. & ZINDER, S. (2001): Assessment of indigenous reductive dechlorinating potential at a TCE-contaminated site using micorocosms, polymerase chain reaction analysis and site data. Environ. Sci. Technol., 35, 1, 1830-1839. FETTER, C. W. (1994): Applied hydrogeology. MacMillan College Publishing, Co. FETZNER, S. (1998): Bacterial dehalogenation. Appl. Microbiol. Biotechnol., 50, 633-657. FRANCIS, C. W. & GRIGAL, D. F. (1971): A rapid and simple procedure using 85Sr for determining cation exchange capacities of soils and clays. Soil Sci., 112, 17-21. FRANK, K. (1991): Tongesteine. Retention von Schwermetallen und die Einflußnahme künstlicher Komplexbildner. Schriftenreihe Angew. Geologie, 11. FREEZE, R. A. & CHERRY, J. A. (1979): Groundwater. Prentice-Hall, Englewood Cliffs, N. J.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
925
FRITSCHE, W. & HOFRICHTER, M. (2000): Aerobic degradation by microorganisms, in Biotechnology, Rehm, H.-J. & Reed, G. (Eds.), Vol. 11b Environmental Processes II, Wiley-VCH, Weinheim. GDA (1993): Empfehlungen des Arbeitskreises „Geotechnik der Deponien und Altlasten“ GDA. 2. Aufl., Deutsche Gesellschaft für Erd- und Grundbau e.V., GDA E 3-3 3; Ernst, Berlin. GEYER, R., PEACOCK, A. D., MILTNER, A., RICHNOW, H. H., WHITE, D. C., SUBLETTE, K. L. & KÄSTNER, M. (2005): In situ assessment of biodegradation potential using biotraps amended with 13C-labeled benzene or toluene. Environ. Sci. Technol., 39, 13, 4983-4989. GIEG, L. M. & SUFLITA, J. M. (2002): Detection of anaerobic metabolites of saturated and aromatic hydrocarbons in petroleum-contaminated aquifers. Environ. Sci. Technol., 36, 17, 3755-3762. GILLHAM, R. W. & O’HANNASIN, S. F. (1994): Enhanced degradation of halogenated aliphatics by zero-valent iron. Groundwater, 32, 6, 958-967. GILLHAM, R. W., ROBIN, M. J. L. & PTACEK, C. J. (1990a): A device for in situ determination of geochemical transport parameters 1. Retardation. Ground Water, 28, 5, 666-672. GILLHAM, R. W., STARR, R. C. & MILLER, D. J. (1990b): A device for in situ determination of geochemical transport parameters 2. Biochemical Reactions. Ground Water, 28, 5, 858-862. GRAF VON REICHENBACH, H. (1966): Anomalien des Kationenaustausches bei Vermiculiten. Z. Pflanzenernährung Bodenkunde, 113, 203. GRBIC-GALIC, D. (1990): Methanogenic transformation of aromatic hydrocarbons and phenols in groundwater aquifers. Geomicrobiol. J., 8, 167-200. GRBIC-GALIC & D. VOGEL, T. M. (1987): Transformation of toluene and benzene by mixed methanogenic cultures. Appl. Environ. Microbiol., 53, 2, 254-260. GRIEBLER, C., SAFINOWSKI, M., VIETH, A., RICHNOW, H. H. & MECKENSTOCK, R. U. (2004): Combined application of stable carbon isotope fractionation for assessing in situ degradation of aromatic hydrocarbons in a tar oil-contaminated aquifer. Environ. Sci. Technol., 38, 2, 617-631. GRIM, R. E. (1962): Applied Clay Mineralogy. McGraw-Hill, New York. HANG, P. T. & BRINDLEY, G. W. (1970): Methylene blue adsorption by clay minerals. Determination of surface areas and cation exchange capacities. Clays Clay Minerals, 10, 203-212. HARRISON, R. M. (Ed.) (1999): Understanding Our Environment. An Introduction to Environmental Chemistry and Pollution. Third Edition. Royal Society of Chemistry. HARTOG, N., GRIFFIOEN, J. & VAN DER WEIJDEN, C. H. (2002): Distribution and reactivity of O2-reducing components in sediments from a layered aquifer. Environ. Sci. Technol., 36, 2338-2344.
926
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
HE, J., SUNG, Y., DOLLHOPF, M. E., FATHEPURE, B. Z., TIEDJE, J. M. & LÖFFLER, F. E. (2002): Acetate versus hydrogen as direct electron donors to stimulate the microbial reductive dechlorination process at chloroethene-contaminated sites. Environ. Sci. Technol., 36, 18, 3945-3952. HEINRICHS, H. & HERRMANN, A. G. (1990): Praktikum der analytischen Geochemie. Springer, Berlin. HENDRICKSON, E., PAYNE, J., YOUNG, R., STARR, M., PERRY, M., FAHNESTOCK, S., ELLIS D. & EBERSOLE, C. (2002): Molecular analysis of Dehalococcoides 16S ribosomal DNA from chloroethene-contaminated sites throughout North America and Europe. Appl. Environ. Microbiol., 68, 2, 485-495. HERON, G. & CHRISTENSEN, T. H. (1995): Impact of sediment-bound iron on redox buffering in a landfill leachate polluted aquifer. Environ. Sci. Technol., 29, 187-192. HILTMANN, W. & STRIBRNY, B. (1998): Handbuch zur Erkundung des Untergrundes von Deponien und Altlasten, Bd. 5, Tonmineralogie und Bodenphysik, Springer, Berlin. HIRSCHORN, S. K., DINGLASAN, M. J., ELSNER, M., MANCINI, S. A., LACRAMPE-COULOUME, G., EDWARDS, E. A., SHERWOOD LOLLAR, B. (2004): Pathway dependent isotopic fractionation during aerobic biodegradation of 1,2-dichloroethane. Environ. Sci. Technol., 3, 18, 4775-4781. HOHNSTOCK-ASHE, A. M., PLUMMER, S. M., YAGER, R. M., BAVEYE, P. & MADSEN, E. L. (2001): Further biogeochemical characterization of a trichloroethene-contaminated fractured dolomite aquifer: electron source and microbial communities involved in reductive dechlorination. Environ. Sci. Technol., 35, 22, 4449-4456. HOPMANS, J. W. & DANE, J. H. (1986): Temperature Dependence of Soil Hydraulic Properties. Soil Sci. Soc. Am. Journal, 50, 1. HOUBEN, G. J. 2003: Iron oxide incrustations in wells. Part 1: genesis, mineralogy and geochemistry. – Applied Geochemistry, 18, 927-939. HOUBEN, G. & TRESKATIS, C. (2006): Rehabilitation of water wells. McGraw Hill, New York. HOURBRON, E., ESCOFFIER, S. & CAPDEVILLE, B. (2000): Trichloroethylene elimination assay by natural consortia of heterotrophic and methanotrophic bacteria. Water Sci. Technol., 42, 5-6, 395-402. HUNKELER, D., ARAVENA, R. & COX, E. (2002): Carbon isotopes as a tool to evaluate the origin and fate of vinyl chloride: laboratory experiments and modeling of isotope evolution. Environ. Sci. Technol., 36, 3378-3384. HUNT, M. J., SHAFER, M. B., BARLAZ, M. A. & BORDEN, R. C. (1997): Anaerobic biodegradation of alkylbenzenes in laboratory microcosms representing ambient conditions. Bioremediation J., 1, 1, 53-64. HUTCHINS, S. R. (1991): Biodegradation of monoaromatic hydrocarbons by aquifer microorganisms using oxygen, nitrate, or nitrous oxide as terminal electron acceptor. Appl. Environ. Microbiol., 57, 8, 2403-2407.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
927
ISO 5667-1 Water quality; Sampling; Part 1: guidance on the design of sampling programmes and sampling techniques. International Organization for Standardization, Geneva. ISO 5667-3 Water quality; Sampling; Part 3: Guidance on the preservation and handling of water samples. International Organization for Standardization, Geneva. ISO 5667-4 Water quality; Sampling; Part 4: guidance on sampling from lakes, natural and man-made. International Organization for Standardization, Geneva. ISO 5667-6 Water quality; sampling; Part 6: guidance on sampling of rivers and streams. International Organization for Standardization, Geneva. ISO 5667-11 Water quality; sampling; Part 11: guidance on sampling of groundwaters. International Organization for Standardization, Geneva. ISO 5667-18 Water quality – Sampling – Part 18: guidance on sampling of groundwater at contaminated sites. International Organization for Standardization, Geneva. ISO 6878 Water quality – Determination of phosphorus – Ammonium molybdate spectrometric method. International Organization for Standardization, Geneva. ISO 8165-2 Water quality – Determination of selected monovalent phenols – Part 2: Method by derivatization and gas chromatography. International Organization for Standardization, Geneva. ISO 8245 Water quality – Guidelines for the determination of total organic carbon (TOC) and dissolved organic carbon (DOC). International Organization for Standardization, Geneva. ISO 9377-2 Water quality – Determination of hydrocarbon oil index – Part 2: Method using solvent extraction and gas chromatography. International Organization for Standardization, Geneva. ISO 9562 Water quality – Determination of adsorbable organically bound halogens (AOX). International Organization for Standardization, Geneva. ISO 10390 Soil quality – Determination of pH. International Organization for Standardization, Geneva. ISO 10694 Soil quality – Determination of organic and total carbon after dry combustion (elementary analysis). International Organization for Standardization, Geneva. ISO 10695 Water quality – Determination of selected organic nitrogen and phosphorus compounds – Gas chromatographic methods. International Organization for Standardization, Geneva. ISO 11074-2 Soil quality – Vocabulary – Part 2: Terms and definitions relating to sampling. International Organization for Standardization, Geneva. ISO 11262 Soil quality – Determination of cyanide. International Organization for Standardization, Geneva. ISO 11272 Soil quality – Determination of dry bulk density. International Organization for Standardization, Geneva.
928
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
ISO 11277 Soil quality – Determination of particle size distribution in mineral soil material – Method by sieving and sedimentation. International Organization for Standardization, Geneva. ISO 11369 Water quality – Determination of selected plant treatment agents – Method using high performance liquid chromatography with UV detection after solid-liquid extraction. International Organization for Standardization, Geneva. ISO 14869-1 Soil quality – Dissolution for the determination of total element content – Part 1: Dissolution with hydrofluoric and perchloric acids. International Organization for Standardization, Geneva. ISO 14869-2 Soil quality – Dissolution for the determination of total element content – Part 2: Dissolution by alkaline fusion. International Organization for Standardization, Geneva. ISO 15009 Soil quality – Gas chromatographic determination of the content of volatile aromatic hydrocarbons, naphthalene and volatile halogenated hydrocarbons – Purgeand-trap method with thermal desorption. International Organization for Standardization, Geneva. ISO 15913 Water quality – Determination of selected phenoxyalkanoic herbicides, including bentazones and hydroxybenzonitriles by gas chromatography and mass spectrometry after solid phase extraction and derivatization. International Organization for Standardization, Geneva. ISO 16703 Soil quality – Determination of content of hydrocarbon in the range C10 to C40 by gas chromatography. International Organization for Standardization, Geneva. ISO 17294-2 Water quality – Application of inductively coupled plasma mass spectrometry (ICP-MS) – Part 2: Determination of 62 elements. International Organization for Standardization, Geneva. ISO 17993 Water quality – Determination of 15 polycyclic aromatic hydrocarbons (PAH) in water by HPLC with fluorescence detection after liquid-liquid extraction. International Organization for Standardization, Geneva. ISO CD 16772 Soil quality – Determination of mercury in aqua regia soil extracts with cold-vapour atomic spectrometry or cold-vapour atomic fluorescence spectrometry. International Organization for Standardization, Geneva. ISO/DIS 5667-1 Water quality – Sampling – Part 1: Guidance on the design of sampling programmes and sampling techniques. International Organization for Standardization, Geneva. ISO/DIS 9377-4 Water quality – Determination of hydrocarbon oil index – Part 4: Method using solvent extraction and gas chromatography. International Organization for Standardization, Geneva. ISO/IEC 17025 General requirements for the competence of testing and calibration laboratories. International Organization for Standardization, Geneva.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
929
ISO/ TR 11046 Soil quality – Determination of mineral oil content – Methods by infrared spectrometry and gas chromatographic method. International Organization for Standardization, Geneva. JASMUND, K. & LAGALY, G. (1993): Tonminerale und Tone. Steinkopff, Darmstadt. JIMÉNEZ, L., ALZAGA, R. & BAYONA, J. M. (2002): Determination of organic contaminants in landfill leachates: a review. Int. J. Environ. Anal. Chem. 82, 7, 415 – 430. JIWAN, J. & GATES, G. (1992): A Practical Guide to Groundwater Sampling, 1st ed. New South Wales Department of Water Resources, Technical Services Division, TS 92 080. JOHANSEN, S. S., HANSEN, A. B., MOSBAEK, H. & ARVIN, E. (1996): Method development for trace analysis of heteroaromatic comounds in contaminated groundwater. J. Chromatogr. A., 738, 295-304. JOHNSTON, C. (1996): Sorption of organic compounds on clay minerals: A surface functional group approach. In: SAWHNEY, B. (Ed.), Organic pollutants in the environment, CMS Workshop Lectures, Vol. 8, The Clay Minerals Society, Boulder, Colorado, 1-44. JUHASZ, A. L. & NAIDU, R. (2000): Bioremediation of high molecular weight polycyclic aromatic hydrocarbons: a review of the microbial degradation of benzo[a]pyrene. Int. Biodeterior. Biodegrad., 45, 1-2, 57-88. JURY W. A., RUSSO D. & SPOSITO, G. (1987): The spatial variability of water and solute transport properties in unsaturated soil: II. Scaling models of water transport. Hilgardia 55, 33-56. KANY, M. (1974): Berechnung von Flächengründungen. 2. Aufl. Ernst, Berlin. KÄSTNER, M. (2000): Degradation of aromatic and polyaromatic compounds, in Biotechnology, REHM, H.-J. & REED, G. (Eds.), Vol 11b Environmental Processes II, Wiley-VCH, Weinheim. KAUFHOLD, S. & DOHRMANN, R. (2003): Beyond the Methylene Blue method: determination of the smectite content using the Cu-triene method. Zs. Angew. Geol., 49, 13-17. KICK, H. (1956): Bemerkungen zur T- und S-Wert-Bestimmung nach Mehlich in Böden mit höheren Gehalten an CaCO3. Z. Pflanzenernährung Düngung Bodenkunde, 75, 67-69. KLEEBERG, R. (2005) Results of the second Reynolds Cup contest in quantitative mineral analysis. IUCr CPD Newsletter, 30, 22-26. KLOSA, D. (1994): Eine rechnergestützte Methode zur Bestimmung des Gesamtkarbonatgehaltes in Sedimenten und Böden. – Z. Angew. Geol., 40, 1, 18-21. KLOTZ, D. & FOLIV, F. (1983): Eine einfache Methode zur Bestimmung der Verteilungskoeffizienten von Radionukliden im Grundwasser. GWF-Wasser/Abwasser 124, 139-141. KÖGEL-KNABNER, I. & TOTSCHE, K. U. (1998): Influence of dissolved and colloidal phase humic substances on the transport of hydrophobic organic contaminants in soils. Phys. Chem. Earth, 23, 2, 179-185.
930
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
KOHLER, E. E. (1985): Mineralogische Veränderungen von Tonen und Tonmineralen durch organische Lösungen. – In: MESECK, H. (Hrsg.): Abdichten von Deponien, Altlasten und kontaminierten Standorten. – Mitt. Inst. Grundbau u. Bodenmechanik TU Braunschweig 20, 87-94. KÖSTER, H. M. (1979): Die chemische Silikatanalyse. Springer, Berlin. KRAUSKOPF, K. B. & BIRD, D. K. (1995): Introduction to Geochemistry. 3rd edn., McGrawHill, New York. KUKKADAPU, R. K., ZACHARA, J. M., FREDRICKSON, J. K., SMITH, S. C., DOHNALKOVA, A. C. & RUSSELL, C. K. (2003): Transformation of 2-line ferrihydrite to 6-line ferrihydrite under oxic and anoxic conditions. – Am. Mineralogist, 88, 11-12, 1903-1914. LAGALY, G. & KÖSTER, H. M. (1993): Tone und Tonminerale. In JASMUND, K. & LAGALY, G. (Ed.): Tonminerale und Tone. Steinkopff, Darmstadt. LAGALY, G. (1985): Clay organic interactions: Problems and results. In: SCHULTZ, L. G., VAN OLPHEN, H. & MUMPTON, F.A. (Eds.): Proc. Int. Clay Conference, Denver, 1985. The Clay Min. Soc., Bloomington, Indiana, 343-351. LAKE, D. L., KIRK, W. W. & LESTER, J. N. (1984): Fractionation, characterization, and speciation of heavy metals in sewage sludge and sludge-amended soils: A Review. LANG, H-J., HUDER, J. & AMANN, P. (1996): Bodenmechanik und Grundbau. 6. Aufl. Springer, Berlin. LANGENHOFF, A. A. M., ZEHNDER, A. J. B. & SCHRAA, G. (1996): Behaviour of toluene, benzene and naphthalene under anaerobic conditions in sediment columns. Biodegradation, 7, 267-274. LANIEWSKI, A. K., DAHLEN, J., BOREN, H. & GRIMVALL, A. (1999): Determination of group parameters for organically bound chlorine, bromine and iodine in precipitation. Chemosphere, 38, 4, 771-782. LECO (1996): CNS-2000 Elemental Analyzer – Instruction Manual. LECO Corp., St. Joseph, MI. LEDUC, R., SAMSON, R., AL-BASHIR, B., AL-HAWARI, J. & CSEH, T. (1992): Biotic and abiotic disappearance of four PAH compounds from flooded soil under various redox conditions. Wat. Sci. Technol., 26, 1-2, 51-60. LEVINE, A. D., LIBELO, E. L., BUGNA, G., SHELLY, T., MAYFIELD, H. & STAUFFER, T. B. (1997): Biogeochemical assessment of natural attenuation of JP-4-contaminated ground water in the presence of fluorinated surfactants. Sci. Total Environ., 208, 179-195. LEWIS, D. R. (1949): Analytical data on reference clay materials. Sect. 3, Base-exchange data. 1950, American Petroleum Institute Project 49 Clay Mineral Standards, Preliminary Report No. 7. LICHT, D., AHRING, B. K. & ARVIN, E. (1996): Effects of electron acceptors, reducing agents, and toxic metabolites on anaerobic degradation of heterocyclic compounds. Biodegradation, 7, 83-90. LIPPMANN, F. (1973): Sedimentary carbonate minerals. Springer, Berlin.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
931
LOAN, M., PIERRE, T. G. S., PARKINSON, G. M., NEWMAN, O. G. M. & FARROW, J. B. (2002). Identifying nanoscale ferrihydrite in hydrometallurgical residues. JOM, 54, 12, 40. LÖFFLER, F. E. & RITALAHTI, K. M. (2005): Bioaugmentation: Insights from the field and application of molecular diagnostic tools. In: DECHEMA, Tagungshandbuch, Perspektiven molekularer und isotopischer Methoden zum Nachweis des natürlichen Schadstoffabbaus in Böden, 29-30 September 2005, Braunschweig, Germany, 51-57. LOVLEY, D. R. (2000): Anaerobic benzene degradation. Biodegradation, 11, 107-116. LOVLEY, D. R., COATES, J. D., WOODWARD, J. C. & PHILLLIPS, E. J. P. (1995): Benzene oxidation coupled to sulfate reduction. Appl. Environ. Microbiol., 61, 3, 953-958. LUA NRW (1994): Determination of PAH in soil samples. Bulletins of the Environmental Agency of the State of Northrhine-Westphalia, 1. Essen, Germany. LUCKNER, L., VAN GENUCHTEN, M. Th. & NIELSEN, D. R. (1989): A Consistent Set of Parametric Models for two-phase flow of immiscible fluids in the subsurface. Water Resources Research, 25, 10, 2187-2193. LUCKNER, L. & SCHESTAKOW, W. M. (1991): Migration Processes in the Soil and Groundwater Zone. Lewis, Boca Raton, FL. LYEW, D., TARTAKOVSKY, B., MANUEL, M.-F. & GUIOT, S. R. (2002): A microcosm test for potential mineralization of chlorinated compounds under coupled aerobic/anaerobic conditions. Chemosphere, 47, 695-699. LYNKILDE, J. & CHRISTENSEN, T. H. (1992): Fate of organic contaminants in the redox zones of a landfill leachate pollution plume (Vejen, Denmark). J. Contamin. Hydrol., 10, 291-307. MALCOLM, R. L. & KENNEDY, V. C. (1969): Rate of cation exchange on clay minerals as determined by specific ion electrode techniques. Soil Sci. Soc Amer. Proc., 34, 247-253. MANAHAN, S. E. (2004): Environmental Chemistry. CRC Press. MANCINI, S. A., ULRICH, A. C., LACRAMPE-COULOUME, G., SLEEP, B., EDWARDS, E. A., & SHERWOOD LOLLAR, B. (2003): Carbon and hydrogen isotopic fractionation during anaerobic biodegradation of benzene. Appl. Environ. Microbiol., 69, 1, 191-198. MARTIN, H., HEIDINGER, M., ERTL, S., EICHINGER, L., TIEHM, A., SCHMIDT, K., KARCH, U. & LEVE, J. (2006): 13C-Isotopenuntersuchungen zur Bestimmung von Natural Attenuation – Abgrenzung und Charakterisierung eines CKW-Schadens am Standort Frankenthal. TerraTech., 3-4, 14-17. MATTES, T. E., COLEMAN, N. V., SPAIN, J. C. & GOSSETT, J. M. (2005): Physiological and molecular genetic analyses of vinyl chloride and ethene biodegradation in Nocardioides sp. strain JS614. Arch. Microbiol., 183, 95-106. MAYMÓ-GATELL, X., CHIEN, Y.-T., GOSSETT, J. M. & ZINDER, S. H. (1997): Isolation of a bacterium that reductively dechlorinates tetrachloroethene to ethene. Science, 276, 1568-1571.
932
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
MCBRIDE, M. B. (1979): An interpretation of cation selectivity variations in M+-M+ exchange on clays. Clays Clay Minerals, 27, 417. MCCARTY, D. K. (2002): Quantitative mineral analysis of clay bearing mixtures: The Reynolds Cup contest. IUCr CPD Newsletter, 27, 12-16. MCFARLAND, M. J. & SIMS, R. C. (1991): Thermodynamic framework for evaluating PAH degradation in the subsurface. Ground Water, 29, 6, 885-896. MCLEAN, E. O. (1982): Soil pH and lime requirement. In: Methods of Soil Analysis, Part 2, 2nd edn., PAGE A. L. et al., Ed., Agronomy., 9, 199-224. Am. Soc. of Agron., Inc., Madison, WI. MCNALLY, D. L., MIHELCIC, J. R. & LUEKING, D. R. (1998): Biodegradation of three- and four-ring polycyclic aromatic hydrocarbons under aerobic and denitrifying conditions. Environ. Sci. Technol., 32, 2633-2639. MCNAUGHT, A. D. & WILKINSON, A. (1997): IUPAC Compendium of Chemical Terminology – The Gold Book. 2nd edn. Blackwell Science. Also available online: www.iupac.org/publications/compendium/index.html MECKENSTOCK, R. U., MORASCH, B., GRIEBLER, C. & RICHNOW, H. H. (2004): Stable isotope fractionation analysis as a tool to monitor biodegradation in contaminated aquifers. J. Contam. Hydrol., 75, 215-255. MECKENSTOCK, R. U., ANNWEILER, E., MICHAELIS, W., RICHNOW, H. H. & SCHINK, B. (2000): Anaerobic naphthalene degradation by a sulfate-reducing enrichment culture. Appl. Environ. Microbiol., 66, 7, 2743-2747. MEHLICH, A. (1948): Determination of cation- and anion-exchange properties of soils. Soil Science, 66, 429-445. MEHRA, O. P, & JACKSON, M. L. (1960): Iron oxide removal from soils and clays by a dithionite-citrate systems buffered with sodium bicarbonate. Clays Clay Minerals, 7, 317-327. MEIER, L. P. & KAHR, G. (1999): Determination of the Cation Exchange Capacity (CEC) of Clay Minerals using the Complexes of Copper (II) Ion with Triethylenetetramine and Tetraethylenepentamine. Clays Clay Minerals, 47, 386-388. MEUNIER, A. (2005): Clays. Springer, Berlin. MEUNIER, A. & VELDE, B. (2004): Illite. Origins, Evolution and Metamorphism. Springer, Berlin. MIDDELDORP, P., LUIJTEN, M., VAN DE PAS, B., VAN EEKERT, M., KENGEN, S., SCHRAA, G. & STAMS, A. (1999): Anaerobic microbial reductive dehalogenation of chlorinated ethenes. Bioremediation Journal, 3, 3, 151-169. MIHELCIC, J. R. & LUTHY, R. G. (1988): Degradation of polycyclic aromatic hydrocarbon compounds under various redox conditions in soil-water systems. Appl. Environ. Microbiol. 54, 5, 1182-1187.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
933
MILLER, M. M., WASIK, S. P., HUANG, G.-L., SHIU, W.-Y. & MACKAY, D. (1985): Relationships between octanol-water partition coefficient and aqueous solubility. Environ. Sci. Technol., 19, 6, 522-529. MOORE, D. M., & REYNOLDS, R. C. (1997): X-ray Diffraction and the Identification and Analysis of Clay Minerals, 2nd edn., Oxford University Press, New York. MORASCH, B. & HUNKELER, D. (2005): Isotopenfraktionierung zur Bestimmung des natürlichen Abbaus von chlorierten und nicht chlorierten Kohlenwasserstoffen. In: DECHEMA, Tagungshandbuch, Perspektiven molekularer und isotopischer Methoden zum Nachweis des natürlichen Schadstoffabbaus in Böden, 29.-30. September 2005, Braunschweig, Germany, 31-38. MORASCH, B., RICHNOW, H. H., Schink, B. & Meckenstock, R. U. (2001): Stable hydrogen and carbon isotope fractionation during microbial toluene degradation: Mechanistic and environmental aspects. Appl. Environ. Microbiol., 10, 4842-4849. MORTLAND, M. M. (1970): Clay organic complexes and interactions. Advances in Agronomy, 22, 75-117. MULL, R., NORDMEYER, H., BOOCHS, P. W. & LIETH, H. (1994): Pflanzenschutzmittel im Grundwasser. Springer, Berlin. MÜLLER, A., SCHÄFER, W., WICKERT, F., TIEHM, A. (2006): Nachweis und Identifikation von Natural Attenuation Prozessen in einer LCKW-Fahne. altlasten spektrum 6/2006: in press. NAGRA (2002): Projekt Opalinuston – Synthese der geowissenschaftlichen Untersuchungsergebnisse – Entsorgungsnachweis für abgebrannte Brennelemente, verglaste hochaktive sowie langlebige mittelaktive Abfälle. Technischer Bericht (NTB), 02-03. NARINE, D. R. & GUY, R. D. (1981): Interactions of some large organic cations with bentonite in dilute aqueous solutions. Clays Clay Minerals, 29, 205-212. NAS-NRC: National Academy of Science, National Research Council (1994): Alternatives for groundwater cleanup. National Academy Press, Washington, D.C.. Nielsen, D. M. (1991): Practical Handbook of Ground-Water Monitoring. Lewis Publishers, Boca Raton, Fl.. NIJENHUIS, I., ANDERT, J., BECK, K., KÄSTNER, M., DIEKERT, G. & RICHNOW, H. H. (2005): Stable isotope fractionation of tetrachloroethene during reductive dechlorination by Sulfurospirillum multivorans and Desulfitobacterium sp. strain PCE-S and abiotic reactions with cyanocobalamin. Appl. Environ. Microbiol., 71, 7, 3413-3419. NITSCHE, C. (1998): Grundwasserprobennahme im Umfeld von Tagebaurestlöchern und – seen. Proceedings des DGFZ e.V., Heft 13, 239-251. NORTHCOTT, G. L. & JONES, K. C. (2000): Spiking hydrophobic organic compounds into soil and sediment: A review and critique of adopted procedures. Environ. Toxicol. and Chem., 19, 2418-2430. NUMATA, M., NAKAMURA, N., KOSHIKAWA, H. & TERASHIMA, Y. (2002): Chlorine isotope fractionation during reductive dechlorination of chlorinated ethenes by anaerobic bacteria. Environ. Sci. Technol., 36, 20, 4389-3494.
934
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
OBERMANN, P. & CREMER, S. (1992): Entwicklung eines Routinetests zur Elution von Schwermetallen aus Abfällen und belasteten Böden. Abschlußbericht Landesamt für Wasser und Abfall NRW von der Ruhruniversität Bochum. OECD (2000): Adsorption – Desorption Using a Batch Equilibrium Method. Organisation for Economic Co-operation and Development (OECD). OECD guideline for the testing of chemicals no. 106. PELZ, O., CHATZINOTAS, A., ANDERSEN, N., BERNASCONI, S. M., HESSE, C., ABRAHAM, W.R. & ZEYER, J. (2001): Use of isotopic and molecular techniques to link toluene degradation in denitrifying aquifer microcosms to specific microbial populations. Arch. Microbiol., 175, 270-281. PLEYSIER, J. & CREMERS, A. (1975): Stability of silver-thiourea complexes in montmorillonite clay. Journal of the Chemical Society-Faraday Transactions I, 256 - 264. PLEYSIER, J. & JUO, A. S. R. (1980): A single-extraction method using silver-thiourea for measuring exchangeable cations and effective CEC in soils with variable charges. Soil Sci., 129, 205-211. POND, K. L., HUANG, Y., WANG, Y. & KULPA, C. F. (2002): Hydrogen isotopic composition of individual n-alkanes as an intrinsic tracer for bioremediation and source identification of petroleum contamination. Environ. Sci. Technol., 36, 724-728. PROVIDENTI, M. A., LEE, H., & TREVORS, J. T. (1993): Selected factors limiting the microbial degradation of recalcitrant compounds. J. Ind. Microbiol., 12, 379-395. PUSCH, R. & YONG, R. N. (2006): Microstructure of Smectite Clays and Engineering Performance. Taylor & Francis, London. QUEVAUVILLE, P. (1996): Harmonization of leaching/extraction tests for environmental risk assessment. Sci. Tot. Environ. 178, 1-132. RAMSAY, J., ROBERTSON, K., MEYLAN, S., LUU, Y.-S., LEE, P. & RAMSAY, B. (2001): Naphthalene and phenanthrene mineralization coupled to Fe(III)-reduction and mechanisms of accessing insoluble Fe(III), in Natural Attenuation of Environmental Contaminants: Proceedings of the 6th Int. In Situ and On Site Bioremediation Symposium, LEESON, A., KELLEY, M. E., RIFAI, H. S. & MAGAR, V. S. (Eds.), vol. 6, 2, Battelle Press, Columbus, OH. REICHERT, J.-K. & ROEMER, M. (1996): Eluatuntersuchungen. In: Fachgruppe Wasserchemie in der GDCh (Hrsg.), Chemie und Biologie der Altlasten. VCH, Weinheim. RETTENBERGER, G & METZGER, H. (1992): Der Deponiegashaushalt in Altablagerungen (Landfill Gas Budget of Abandoned Waste Sites). In: Handbuch Altlasten (Manual for Hazardous Sites). Landesanstalt für Umweltschutz Baden-Württemberg (State Institute of Environmental Protection, Baden Württemberg), Germany. ROCKNE, K. J., CHEE-SANFORD, J. C., SANFORD, R. A., HEDLUND, B. R., STALEY, J. T. & STRAND, S. E. (2000): Anaerobic naphthalene degradation by microbial pure cultures under nitrate-reducing conditions. Appl. Environ. Microbiol., 66, 4, 1595-1601.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
935
RÖLING W. F. M. & VAN VERSEVELD H. W. (2002): Natural attenuation: What does the subsurface have in store? Biodegradation, 13, 53-64. RUMP, H.-H. & SCHOLZ, B. (1995): Untersuchung von Abfällen, Reststoffen und Altlasten (Investigation of wastes, residues, and hazardous sites). Wiley-VCH, Weinheim. SAFINOWSKI, M., GRIEBLER, C. & MECKENSTOCK, R. U. (2006): Anaerobic cometabolic transformation of polycyclic and heterocyclic aromatic hydrocarbons: evidence from laboratory and field studies. Environ. Sci. Technol., 40, 13, 4165-4173. SAGNER, A., BRINKMANN, C., EISENTRÄGER, A., HINGER, G., HOLLERT, H. & TIEHM, A. (2006): Vorkommen und Ökotozität von heterozyklischen Kohlenwasserstoffen (NSOHET). In: Tagungshandbuch des BMBF-Workshops zum KORA Themenverbund 2: “Gaswerke, Kokereien, Teerverarbeitung”: MNA bei der Altlastenbehandlung, Duisburg 12. Juni 2006 und Dresden 19. Juni 2006, 37-41. SAGNER, A. & TIEHM, A. (2005): Enhanced natural attenuation of heterocyclic hydrocarbons: biodegradation under anaerobic conditions and in the presence of H2O2. In: UHLMANN, O. & ANNOKKÉE, F. (Eds.): Proceedings (CD) of the 5th international FZK/TNO conference on soil-water systems, 03-07 October, Bordeaux, France, 1629-1636. SCHEIDEGGER, A., BÜRGISSER, C. S., BORKOVEC, M., STICHER, H. MEEUSSEN & VAN RIEMSDIKK, W. H. (1994): Convective Transport of acids and bases in porous media. Water Resour. Res., 30, 2937-2944. SCHINK, B. (2000): Principles of anaerobic degradation of organic compounds, in Biotechnology, REHM, H.-J. & REED, G. (Eds.), vol. 11b Environmental Processes II, Wiley-VCH, Weinheim. SCHMIDT, K.R., STOLL, C. & TIEHM, A. (2006a): Evaluation of 16S-PCR detection of Dehalococcoides at two chloroethene-contaminated sites. Water Sci. Technol., in press. SCHMIDT, K. R., STOLL, C. & TIEHM, A. (2006b): 16S-PCR detection of halorespiring bacteria at two chloroethene-contaminated sites in Germany. In: UFZ Centre for Environmental Research Leipzig-Halle, Book of Abstracts, International Conference on Environmental Biotechnology, 09-14 July, Leipzig, Germany, 233. SCHREINER, M., & KREYSING, K. (1998): Handbuch zur Erkundung des Untergrundes von Deponien und Altlasten, Bd. 4, Geotechnik Hydrogeologie, Springer, Berlin. SCHRIEVER, M. & HIRNER, A. (1994): Entwicklung von Routinetests zur Elution von organischen Komponenten aus Abfällen und belasteten Böden. Abschlußbericht Landesamt für Wasser und Abfall NRW von der Universität GH Essen. SCHULZ, H.-D. & KÖLLING, M. (1992): Grundlagen und Anwendungsmöglichkeiten hydrogeochemischer Modelle. In: DVWK (Hrsg.): Anwendung hydrogeochemischer Modelle.- DVWK-Schriften 100, Parey, Berlin. SCHULZE, S. & TIEHM, A. (2004): Assessment of microbial natural attenuation in groundwater polluted with gasworks residues. Water Sci. Technol., 50, 5, 347-353.
936
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
SCHULZE, S., BIRKLE, M., EBNER, R., BARCZEWSKI, B. & TIEHM, A. (2003): Natural microbial degradation in a BTEX/PAH polluted groundwater plume. In: Proceedings of ConSoil 2003 (CD), 8th Int. FZK/TNO Conf. Contamin. Soil, 12.-16.05.2003 in Gent, Belgium; Theme C, 2212-2218. SCHUMACHER, B. A., NEARY, A. J., PALMER, C. J. et al. (1995): Laboratory Methods for Soil and Foliar Analysis in Long-Term Environmental Monitoring Programs. EPA/600/R-95/077. U.S. EPA, Las Vegas. SCHUMACHER, B. A. (2002): Methods for the determination of total organic carbon (TOC) in soil and sediments. United States Environmental Protection Agency, Environmental Sciences Division National, Exposure Research Laboratory Las Vegas, NCEA-C- 1282, www.epa.gov/esd/cmb/research/papers/bs116.pdf. SCHUTH, C., TAUBALD, H., BOLANO, N. & MACIEJCZYK, K. (2003): Carbon and hydrogen isotope effects during sorption of organic contaminants on carbonaceous materials. J. Contam. Hydrol., 64, 3-4, 269-281. SCHWERTMAN, U. & MURAD, E. (1983): Effect of pH on the formation of goethite and hematite from ferrihydrite. Clays and clay minerals, 31, 277-284. SEARLE, P. L. (1986): The measurement of soil cation exchange properties using the single extraction, silver-thiourea method: An evaluation using a range of New Zealand soils. Australian J. of Soil Res., 24, 193-200. SEEL, F. (1979): Grundlagen der analytischen Chemie unter besonderer Berücksichtigung der Chemie in wäßrigen Systemen. (Basics of analytical chemistry with a particular focus on chemistry of aqueous systems). 7th edn. VCH, Weinheim. SHERWOOD LOLLAR, B., SLATER, G. F., SLEEP, B., WITT, M., KLECKA, G. M., HARKNESS, M. & SPIVACK, J. (2001): Stable carbon isotope evidence for intrinsic bioremediation of tetrachloroethene and trichloroethene at Area 6, Dover Air Force Base. Environ. Sci. Technol., 35, 2, 261-269. SHIMADZU (2006): Shimadzu's TOC-V Series. www.ssi.shimadzu.com/products/product.cfm?product=visionarytoc SIMS, R. C. & OVERCASH, M. R. (1983): Fate of polynuclear aromatic compounds (PNAs) in soil plant systems. Residue Rev., 88, 1-68. SIMUNEK, J., SEJNA, M.. & VAN GENUCHTEN, M. T. (1998): The HYDRUS-1D software package for simulating the one-dimensional movement of water, heat, and multiple solutes in variably-saturated media. US Salinity Laboratory, Agricultural Research Service, US Department of Agriculture, Riverside, CA. SIMUNEK, J., VAN GENUCHTEN, M. T., SEJNA, M., TORIDE, N. & LEIJ, F. J. (1999): The STANMOD Computer Software for Evaluating Solute Transport in Porous Media Using Analytical Solutions of the Convection-Dispersion Equation, U.S. Salinity Laboratory, USDA/ARS, Riverside, CA. SINGH, H., LÖFFLER, F. E. & FATHEPURE, B. Z. (2004): Aerobic biodegradation of vinyl chloride by a highly enriched mixed culture. Biodegradation, 15, 197-204.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
937
SLATER, G. F., SHERWOOD LOLLAR, B., SLEEP, B. E. & EDWARDS, E. A. (2001): Variability in carbon isotopic fractionation during biodegradation of chlorinated ethenes: Implications for field applications. Environ. Sci. Technol., 35, 5, 901-907. SMALL, H. (1989): Ion Chromatography. Plenum Press, New York. SÖHNGEN, K. (1992): Strategie einer sach- und fachgerechten Entnahme von Grundwasserproben. eretec. Institut für chemische Analytik und Umwelttechnik, Gummersbach, Germany. SONG, D. L., CONRAD, M. E., SORENSON, K. E. & ALVAREZ-COHEN, L. (2002): Stable carbon isotope fractionation during enhanced in situ bioremediation of trichloroethene. Environ. Sci. Technol., 36, 2262-2268. STANGER, G. (1994): Dictionary of hydrology and water resources. Lochnan, Adelaide. STEINMANN, P. & SHOTYK, W. (1996): Sampling anoxic pore waters in peatlands using “peepers” for in-situ filtration. Fresenius’ Journal of Analytical Chemistry, 354, 709-713. STIEBER, M., HAESELER, F., WERNER, P. & FRIMMEL, F. H. (1994): A rapid screening method for micro-organisms degrading polycyclic aromatic hydrocarbons in microplates. Appl. Microbiol. Biotechnol., 4, 753-755. STRAATEN, L. & HESSER, F. (1998): Grundwasserprobennahme. In: VOIGT, H.-J. & WIPPERMANN, Th., Geochemie, Bd. 6 des Handbuch zur Erkundung des Untergrundes von Deponien und Altlasten, Springer, Berlin.
VAN
STRECK, T., POLETIKA, N., JURY, W. A. & FARMER, W. J. (1995): Description of simazine transport with rate – limited, two-stage, linear and nonlinear sorption. Water Resources Research, 31, 811-822. STRECK, T. & RICHTER, J. (1997): Heavy metal displacement in a sandy soil at the field scale: I. Measurements and parameterization of sorption. J. of Environ. Qual., 26, 49-56. STRECK, T. & RICHTER, J. (1999): Field-scale study of chlortoluron movement in a sandy soil over winter: II. Modeling. J. Environ. Qual., 28, 1824-1831. SUAREZ, M. P. & RIFAI, H. S. (1999): Biodegradation rates for fuel hydrocarbons and chlorinated solvents in groundwater. Bioremediation Journal, 3, 4, 337-362. TADJERPISHEH, N. & ZIECHMANN, W. (1986): Genese und Analyse von Ton-HuminstoffKomplexen. – Mitt. Dtsch. Bodenkdl. Gesellsch. 45, 155-160. TAKEUCHI, M., NANBA, K., IWAMOTO, H., NIREI, H., KUSUDA, T., KAZAOKA, O., OWAKI, M. & FURUYA, K. (2005): In situ bioremediation of a cis-dichloroethylenecontaminated aquifer utilizing methane-rich groundwater from an uncontaminated aquifer. Water Res., 39, 2438-2444. TESSIER, A., CAMPBELL, P. G. C. & BISSON, M. (1979): Sequential extraction procedure for the speciation of particulate trace metals. Anal. Chem., 51, 844-851. THENG, B. K. G. (1974): The chemistry of clay organic reactions. Hilger, London/ Wiley, New York.
938
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
THENG, B. K. G. (1979): Formation and properties of clay-polymer complexes. Elsevier, Amsterdam. TIEHM, A. & SCHULZE, S. (2003): Intrinsic aromatic hydrocarbon biodegradation for groundwater remediation. Oil & Gas Science and Technology – Rev. IFP, 58, 4, 449-462. TIEHM, A., GOZAN, M., MÜLLER, A., SCHELL, H., LORBEER, H. & WERNER, P. (2002a): Sequential anaerobic/aerobic biodegradation of chlorinated hydrocarbons in activated carbon barriers. Water Sci. Technol.: Water Supply, 2, 2, 51-58. TIEHM, A., SCHMIDT, K. R., STOLL, C., MÜLLER, A., LOHNER, S., HEIDINGER, M., WICKERT, F. & KARCH, U. (2006a): Assessment of natural microbial dechlorination. Ital. J. of Engin. Geol. and Envir., in press. TIEHM, A., SCHMIDT, K., STOLL, C., MÜLLER, A. & LOHNER, S. (2005): Natürlicher mikrobieller Abbau (Natural Attenuation) von CKW: Fallbeispiele, Abbaumechanismen und Nachweismethoden. In: Ressourcen- und Grundwasserschutz, Veröffentlichungen aus dem Technologiezentrum Wasser (ISSN 1434-5765), Bd. 28. 53-73. TIEHM, A., SCHMIDT, K. R., MARTIN, H. & HEIDINGER, M. (2006b): Stable isotope fractionation during PCE halorespiration and aerobic cisDCE and VC biodegradation. In: UFZ Centre for Environmental Research Leipzig-Halle, Book of Abstracts, International Conference on Environmental Biotechnology, 09-14 July, Leipzig, Germany, 408. TIEHM, A. & FRITZSCHE, C. (1995): Utilization of solubilized and crystalline mixtures of polycyclic aromatic hydrocarbons by a Mycobacterium sp. Appl. Microbiol. Biotechnol., 42, 964-968. TIEHM, A. & STIEBER, M. (2001): Strategies to improve PAH bioavailability: Addition of surfactants, ozonation and application of ultrasound. In: STEGMANN, R., BRUNNER, G., CALMANO, W. & MATZ, G. (Eds.): Treatment of Contaminated Soil, Springer, Berlin. TIEHM, A., SCHULZE, S. & MÜLLER, A. (2002b): Handlungsoption „Natural Attenuation“ – Natürlicher Abbau von Schadstoffen im Grundwasser. In: Aktuelle Themen bei der Trinkwassergewinnung, Veröffentlichungen aus dem Technologiezentrum Wasser (ISSN 1434-5765), Bd. 18, 65-78. TIEHM, A., STIEBER, M., WERNER, P. & FRIMMEL, F. H. (1997): Surfactant-enhanced mobilization and biodegradation of polycyclic hydrocarbons in manufactured gas plant soil. Environ. Sci. Technol., 31, 2570-2576. TOLMAN, C. F. (1937): Groundwater. McGraw-Hill, New York. TORIDE, N., LEIJ, F. J. & VAN GENUCHTEN, M. T. (1995): The CXTFIT Code for Estimating Transport Parameters from Laboratory or Field Tracer Experiments, Version 2.0. Research Report No. 137, U.S. Department of Agriculture, Riverside, CA. TORSTENSSON, B. A. & PETSONK, A. M. (1988): A hermetically isolated sampling method for groundwater investigations. ASTM American Special Technical Publication 963, 274-289.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
939
TOTSCHE, K. U. (2001): Reaktiver Stofftransport in Böden: Optimierte Experimentdesigns zur Prozeßidentifikation. Bayreuther Bodenkundl. Ber., 75. TRIBUTH, H. & LAGALY, G. (1986a): Aufbereitung und Identifizierung von Boden- und Lagerstättentonen I. – GIT Fachz. Lab., 6, 524-529. TRIBUTH, H. & LAGALY, G. (1986b): Aufbereitung und Identifizierung von Boden- und Lagerstättentonen II. – GIT Fachz. Lab., 8, 771-776. UIT (2000): Druckhaltende Schöpfer. Umwelt- und Ingenieurtechnik GmbH, Dresden, Germany. URE, A. M. & DAVIDSON, C. M. (1995): Chemical speciation in the environment. Blackie, London. USEPA (1994): Terms of environment, glossary, abbreviations, and acronyms. U.S. Environmental Protection Agency EPA 175-B 94-015. VERCE, M. F., GUNSCH, C. K., DANKO, A. S. & FREEDMAN, D. L. (2002): Cometabolism of cis-1,2-dichloroethene by aerobic cultures grown on vinyl chloride as the primary substrate. Environ. Sci. Technol., 36, 10, 2171-2177. VINOGRADOV, A. P. (1954): Geochemie seltener und nur in Spuren vorhandener chemischer Elemente in Böden. Akademie Verlag, Berlin. VOIGT, H.-J. & WIPPERMANN, T. (1998): Handbuch zur Erkundung des Untergrundes von Deponien und Altlasten, Bd. 6, Geochemie. Springer, Berlin. WALKER, G. F. (1959): Diffusion of exchangeable cations in vermiculite. Nature, 184, 1392-1393. WALKLEY, A. & BLACK, I. A. (1934): An Examination of Degtjareff Method for Determining Soil Organic Matter and a Proposed Modification of the Chromic Acid Titration Method. Soil Sci., 37, 29-37. WEISS, J. (1991): Ionenchromatographie. 2. erw. Aufl., VCH, Weinheim. WELZ, B. & SPERLING, M. (1997): Atomabsorptionsspektrometrie (Atomic absorption spectrometry). 4th edn. Wiley-VCH, Weinheim. WHO (1998): International Programme on Chemical Safety (IPCS); Environmental Health Criteria 202: Selected Non-heterocyclic Aromatic Hydro-carbons. World Health Organization. Wissenschaftliche Verlagsgesellschaft, Stuttgart. WIEDEMEIER, T. H., RIFAI, H. S., NEWELL, C. J. & WILSON, J. T. (1999): Natural Attenuation of Fuels and Chlorinated Solvents in the Subsurface. Wiley & Sons, New York. WILLIAMS, I. (2001): Environmental Chemistry: A Modular Approach. Wiley & Sons, New York. WILSON, W. E. & MOORE, J. E. (Eds.) (2001): Glossary of Hydrology. American Geological Institute.
940
5.3 Methods for Characterizing the Geochemical and Microbiological Conditions
WINKLER, A. (1989) Untersuchungen zur Mobilität von Technetium (und Selen) in norddeutschen Grundwasserleitern und Technetium im Kontakt mit natürlich vorkommenden Mineralien., Berliner geowiss. Abh., 117. VAN DER ZEE,
S. E. A. T. M. & VAN RIEMSDIJK, W. H. (1987): Transport of Reactive Solute in Spatially Variable Soil Systems. Water Resources Research, 23, 11, 2059-2069.
ZEIEN, H. & BRÜMMER, G. W. (1991): Chemische Extraktionen zur Bestimmung von Schwermetallbindungsformen in Böden. Mitt. Dt. Bodenk. Ges. 59/1. ZHENG, C. & BENNETT, G. D. (1995): Applied contaminant transport modeling: theory and practice. Van Nostrand Reinhold, New York. ZURMÜHL, T. (1994): Validierung konvektiv-dispersiver Modelle zur Berechnung des instationären Stofftransports in ungestörten Bodensäulen. Bayreuther Bodenkundliche Berichte 36.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
941
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results FLORIAN JENN, CLAUS KOFAHL, MIKE MÜLLER, JENS RADSCHINSKI & HANS-JÜRGEN VOIGT
5.4.1 Statistical Methods FLORIAN JENN & HANS-JÜRGEN VOIGT
Almost all data obtained in geoscientific investigations need some kind of statistical treatment for interpretation, as well as for the assessment of reliability and errors. The following statistical methods are frequently used to analyze data sets: x univariate analysis, i.e., analysis of an individual parameter (e.g., concentration of a substance, pH, or electrical conductivity at several locations), x multivariate analysis, i.e., analysis of several parameters together, to determine the relationship between parameters (e.g., the relationship between concentrations of different substances and/or of several environmental parameters), and x time series analysis (e.g., analysis of a parameter as a function of time, for example, monitoring of water level or the concentration of a substance in a groundwater observation well). Geostatistics, interpolation of spatial data, and tests specific for hydrogeochemical data are used to evaluate geological, hydrogeological and geochemical data. Commonly used statistical software are, for example, Origin, SPSS, S-Plus and R. Simple statistical evaluations can also be performed using spreadsheet software (e.g., Microsoft Excel, OpenOffice Calc). Various aspects and examples of interpretation using statistical methods are described in Part 6 of this handbook. Measurements are usually made to determine or estimate the typical (representative, average) values and the range of variation of a parameter (e.g., hydraulic conductivity) of a geological unit. All values of a parameter, not only those actually observed, but those that are potentially observable, are called a statistical population (SWAN & SANDILANDS, 1995). The population is, therefore, established by the choice of the investigation target.
942
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
It is, of course, not feasible to carry out all those possible measurements. Instead, a limited set of measurements is carried out. In statistics, the individual measurements are commonly called observations. Taken together, they comprise a subset of the population, the statistical sample, which is the data available for statistical analysis. The statistical and the geological meanings of the word “sample” must not be confused. A sample in geological terms is, for example, a piece of rock or a quantity of water drawn from a well, and yields a single value for any one property. Several measurements of that property comprise a statistical sample, which can be analyzed to estimate other properties, for example the hydraulic conductivity of a geological unit (e.g., an aquifer) or the concentration of a substance in the aquifer.
5.4.1.1 Univariate Statistics Statistical analysis is commonly carried out stepwise. The first step in the statistical treatment of a data set is a univariate analysis of the data, i.e. each variable (e.g., water level, concentration of substances, layer thickness) in a data set is analyzed separately. The aim of this kind of analysis is to describe the properties of the population by analyzing the sampled data. In the following, the selected variable is denoted X, with the observed values x1,…,xn, where n is the sample size. Histogram and frequency distribution Histograms are used to visualize the frequency distribution of the given values xi. The range of observed values is divided into (usually equally sized) classes. A rectangle is drawn for each class with a width from the lower to the upper boundary of the class and with a height corresponding to the number or percentage of values xi observed within the class. An example is given in Figure 5.4-1. When the number of classes nc is chosen, a compromise has to be found between too few classes, which may suppress information, and too many classes, which causes the variation in the individual values to have too much influence. As a starting point, the statistical software may suggest a number of classes, or the following rules of thumb may be used. Number of classes nc according to DVKW (1990): nc nc
for n < 1000 10 lg n for n > 1000. n
(5.4.1a)
Number of classes nc according to Sturges’ Rule (EVERITT, 1998): nc
log 2 n 1 .
(5.4.1b)
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
943
Fig. 5.4-1: Example of a histogram
Between 5 and 20 classes are usually used (DVKW, 1990). A histogram shows the frequency distribution of the statistical sample, which approximates the frequency distribution of the population. The latter is often drawn as a smooth curve instead of a histogram, and sometimes described in an idealized way by mathematical probability functions (described in more detail below). The frequency distribution normally shows one or more distinct peaks, called modes. These indicate values for x that have been observed most often. Depending on the number of modes, the distribution is called unimodal (one mode), bimodal (two modes) or multimodal (more than two modes). If observations are taken from a single population (e.g., chloride concentration in groundwater samples from various parts of an aquifer) the resulting distribution is usually unimodal. Several modes often indicate that several populations are superimposed (e.g., chloride concentration in samples from an uncontaminated and a contaminated region of an aquifer). Summary statistics and properties of the population There are several parameters to describe the properties of a statistical sample and the corresponding population. Although they could be computed for any data set, they usually only make sense for unimodal data sets. The most important statistical parameters are those for average or typical values for the sample or the population. The sample mean is defined as x
1 n ¦ xi n i=1
(5.4.2)
and is an estimate of the population mean µ. The median is the value in an ordered set of values for which there is an equal number of values larger and
944
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
smaller than it. The mode is the most frequently occurring value in a data set. In a symmetrical, unimodal distribution, the values for these three parameters are identical. Asymmetric, or skewed, distributions are often found (Fig. 5.4-1). This may be caused by a general characteristic of the population (natural concentrations of trace substances are usually steep on the left side and have a longer tail on the right side) or by outliers (individual very high or low values). If the histogram is steep on the left-hand side gradual slopes to the right, it is called positively skewed; if it is steep on the right-hand side, it is called negatively skewed. In skewed distributions, the median and mean move in the direction of the tail with respect to the mode: mode < median < mean, if positively skewed
(5.4.3a)
mean < median < mode, if negatively skewed.
(5.4.3b)
This also shows that the mean is more strongly affected by asymmetry and outliers than the median. On the other hand, the mean is easier to deal with mathematically, and many statistical methods and tools are based on it. Another group of parameters describes the dispersion, or spread, of a distribution, i.e. the amount by which a set of observations deviate from their average value. The most simple is the range, the difference between the maximum and minimum values observed. This, however, is very sensitive to outliers. A more robust approach involves quartiles or percentiles: The first quartile is the value below which the lowest quarter of the sorted set of observations can be found. The third quartile is the value above which is the highest quarter of the observations. The second quartile is the median. The range between first and third quartile is called the interquartile range (IQR) and comprises half of the samples – the ones that are closest to the center. Percentiles are defined in an analogous way. For example, the 5th percentile is the value below which the lowest 5 percent of observations are found. The range between 5th and 95th percentiles is often used as a range of representative values. The observations outside of this range are considered outliers. Another way to characterize spread is to quantify, on average, how far each observation is from the center. The sample variance s2 is thus defined as the average of the squared deviations of each observation xi from the mean x : s2
1 n ¦ x x n 1 i=1 i
2
(5.4.4)
and the standard deviation s is the square root of the variance. These parameters can be used to roughly describe a set of observations, e.g., by giving mean and standard deviation, or the so-called five-number summary (minimum, 1st quartile, median, 3rd quartile, maximum). The latter is the basis for box-and-whisker plots (also called boxplots for short), of which several variations are in use. The basic principle is to draw a box extending
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
945
from the 1st to 3rd quartile (representing the IQR), intersected by a line at the median. Two lines (“whiskers”) extend from the box to one of the following: x the minimum and maximum values, x selected percentiles (e.g., 5th and 95th), or x the furthest observation within a distance of 1.5 IQR from the box (considering anything further away at least a “mild” outlier). This design can be extended to include additional information: outliers can be indicated separately, the mean shown by an additional dot, or notches to indicate medians in other distributions that are significantly different. Figure 5.4-2 shows a sample distribution and the resulting basic boxplot. The boxplot is an alternative to the histogram and is a condensed view of the sample distribution. It quickly shows the median, possibly the mean, and the range of typical values (50 % of the observations are within the box), and indicates the spread of the distribution by the length of the box and the length of the whiskers. Skewness can be seen by off-center placement of the median line as well as different lengths of the “whiskers”. However, it is not possible to discern the modality of the distribution, i.e. whether multiple independent populations are superimposed in the sample (Fig. 5.4-3).
Fig. 5.4-2: Histogram of the frequency distribution and corresponding boxplot containing the line for the median. Circles indicate outliers, the black dot is the mean. “Whiskers” stretch to the furthest observations no more than 1.5·IQR away.
946
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
Fig. 5.4-3: Histogram of a multimodal distribution and corresponding boxplot (same design as in Fig. 5.4-2)
Probability distributions The probability density function f(x) of a continuous random variable X is a statistical function that shows how the density of possible observations in a population is distributed. It is the mathematical ideal corresponding to the empirical frequency distribution of samples described above. The probability density function is the derivative of the cumulative distribution function F(x) if F(x) is differentiable. It can never be negative and its total integral is unity. Geometrically, f(x) is the ordinate of a curve such that f(x) dx yields the probability that the variable will assume some value within the range dx of x. For each differentiable probability density function, there is a cumulative probability distribution F x
x
³ f x dx ,
(5.4.5)
f
which describes the probability of X d x. Values of F(x) for different distributions are available as tables or from statistical software. The cumulative probability distribution can be used to calculate the probability that X falls between two values x1 and x2: P x1 d X d x 2
F x 2 F x1
(5.4.6)
There are several mathematical functions that describe such probability distributions. The most important is the normal distribution, which is a
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
947
symmetrical, bell-shaped curve (Fig. 5.4-4). Given the distribution mean µ and the standard deviation V, the distribution is described by 1
f N x
V 2S
e
x P 2 2V 2
.
(5.4.7)
The normal distribution has been found to describe remarkably well the errors of observation in physics; many variables measured in environmental investigations are (approximately) normally distributed (WEBSTER & OLIVER, 2001). Also, many statistical methods are based on the assumption of data with a normal distribution. If µ = 0 and V = 1, the result is called the standard normal distribution, which is the only normal distribution that is usually tabulated. If a variable X has a normal distribution, the corresponding standardized variable Z
X P
V
(5.4.8)
has a standard normal distribution. Conversely, values of any normal distribution can be calculated from the standard normal distribution by rearranging Equation (5.4.8) X
VZ P.
Fig. 5.4-4: Probability density of the standardized normal distribution
(5.4.9)
948
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
Another important distribution is the log-normal distribution. This is the distribution of a variable whose logarithm has a normal distribution. Given the probability density function of the normal distribution fN, the log-normal distribution is defined as: f LN x
1 xV 2S
e
ln x P 2 2V 2
.
(5.4.10)
This distribution is not symmetrical, but positively skewed. A typical example of a variable with a log-normal distribution is the concentration of a trace substance. There are many more distributions for various situations and purposes, e.g. Student’s t distribution, F distribution, and F2 distribution for statistical inference (see below), which are not described here. Example: Determining natural groundwater conditions by component separation analysis The frequency distribution of concentrations in groundwater was used to determine the natural groundwater quality in aquifers in Germany (WENDLAND et al., 2005). They assume that the histograms can be modeled by a log-normal distribution for the natural concentration distribution and a superimposed normal distribution reflecting local or anthropogenic influences (Fig. 5.4-5a). Using the software package Origin, the frequency distributions of the observed data were separated. Then the 10th and 90th percentiles (i.e. comprising 80 % of values) were taken as the range of natural groundwater conditions (Fig. 5.4-5b). Table 5.4-1 shows the natural groundwater conditions determined by this technique for the main hydrostratigraphic units of major interest to water management.
(a)
(b)
Fig. 5.4-5: Component separation analysis of groundwater data from Germany, WENDLAND et al. (2005). (a) Assumption of two superimposed distributions, (b) percentiles as the range of natural groundwater conditions
949
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
Determination of natural groundwater quality is not only a basis for assessing the influence of various contamination sources, but also for decisions on remediation goals. In this case, however, the possible influence of groundwater upstream from a contamination source, for example a landfill, has to be kept in mind. Table 5.4-1: Results of assessment of natural groundwater conditions in Germany using component separation analysis, WENDLAND et al. (2005) Sands and gravel of aquifer F2 (Saale glaciaton)
conductivity
Jurassic limestone (Malm)
Triassic Triassic limestone sandstone (Muschelkalk) (Buntsandstein)
from
to
from
to
from
to
from
to
µS cm
186
521
387
704
637
939
50
256
-1
0.2
4.6
6
11
3
10
5
11
6.0
7.8
7.1
7.7
7.0
7.5
6.8
7.7
-1
O2
mg L
pH
–
DOC
mg L-1
0.8
5.0
0.3
1.3
0.4
1.2
0.3
1.6
Ca
mg L-1
29
143
69
126
99
154
7
29
Mg
-1
mg L
3
30
4
37
17
50
2
23
Na
mg L-1
6
24
1.3
6.3
3.0
9.2
2
16
K
mg L-1
0.8
4.0
0.3
1.9
0.6
2.1
1.3
3.6
NH4
mg L-1
1.0 mg L-1 2+ Ca + Mg2+ > 1.0 mg L-1
962
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
Fig. 5.4-12: Scatter plot of ionic strength I versus electrical conductivity ECmeas. The dashed line shows the ideal relationship described by Equation (5.4.17). Most samples, especially those with elevated mineral content, fall short of the I value they should have according to EC measurements, indicating that the measured concentrations of some ions might be too low or that some ion was not analyzed at all.
There are more sophisticated procedures for calculating the electrical conductivity from chemical analysis data that could be employed in plausibility testing. In most cases, the method of ROSSUM (1975) works very well.
5.4.2 Conceptual Model HANS-JÜRGEN VOIGT & JENS RADSCHINSKI
A conceptual model is a mental image of an object, system, or process. It consists normally of a simplified, schematic written description and a visual representation. Conceptual hydrogeological models have been used for several decades to describe and understand hydrogeological systems. Such conceptual models are more concerned with the physical than the chemical aspects of the natural system. During the last few decades conceptual models of investigated sites have come into frequent use to assess ecosystem features and processes (including biological, physical, chemical and geomorphic components) of an environment (e.g., new or abandoned landfills, industrial site or mining site).
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
963
A conceptual model is developed as part of the site investigation and reflects the progress of site characterization. Therefore, the conceptual model is the basis as well as the final result of site characterization. The conceptual model is not a fixed rigid idea, but a flexible concept (ORESKES & BELITZ, 2001) which is upgraded iteratively on the basis of the information obtained during the field and laboratory investigations. The first step in a site investigation is to define together with the client the objectives, the boundaries of the site, and the scale. A preliminary conceptual model is prepared on the basis of the orientating investigation of the site. This preliminary conceptual model is the basis for planning the geophysical, geological, hydrogeological, geochemical and microbiological investigations. The product of the site investigation is a verified conceptual model. This is a sound basis for site assessment and all remediation measures. Experience in Germany shows that remedial measures that are not based on a conceptual model are not effective or failed completely in almost 80 % of the cases. The objectives and the sequence of site investigations are outlined in Part 1 of this handbook. Some aspects must be repeated in this section for the discussion of the development of a conceptual model. Site investigations usually take place in the following progression: x orientating site investigation, x preliminary conceptual model, x field survey for a detailed site investigation, x refining of the conceptual model, x verified or final conceptual model, and x risk assessment. The refined conceptual model can be the basis for numerical modeling of groundwater flow and contaminant transport. Numerical modeling is one way to verify the conceptual model. Table 5.4-4 gives elements of a conceptual site model. The hydrogeological model of a site is the nucleus of every conceptual site model. Therefore, the development of the conceptual model should start with an analysis of the hydrogeological system: x the spatial distribution of hydrostratigraphic units and their lithological and petrophysical properties, x groundwater flow, and x hydraulic boundary conditions.
characterization of a site for a new landfill, industrial site or mine site (e.g., spoil/slag heap, ore treatment plant) inspection of an operating landfill, industrial site or mine site (e.g., spoil/slag heap, ore treatment plant)
main objective
base data
topography, land use and vegetation, settlements, roads and railways
mining site (e.g., spoil/slag heap, ore treatment plant) - locally used site name; - owner/user of the site; - geographical coordinates (e.g., determined by GPS); - sheet number of topographical map; - site plan; - size of the site (in m2); - classification of the site suspected to be hazardous (see above); - use of the site in the past and at present, e.g., type of industrial operation; - operation period for industrial sites, landfills or mining sites; - contamination expected on the basis of the type of industrial operation; - in the case of a landfill, type of waste deposition (e.g., stockpiling, heap on a slope, backfill of a quarry, backfill of a gravel pit), type of material deposited (e.g., uncontaminated excavated soil, natural rock, minerals, municipal waste, industrial waste), type of contaminants expected (e.g., water-soluble substances, gaseous and volatile substances, inorganic substances, organic substances, toxic, carcinogenic, mutagenic substances, warfare agents, explosives, radioactive materials); and - person(s) responsible for releases of hazardous substances at the site
characterization of a site suspected to be hazardous waste deposit (landfill, sewage sludge deposition, small site with buried and/or heaped waste classification of sites suspected to be and litter) hazardous industrial site (e.g., galvanizing plant, metallurgical workshop, mechanical workshop, motor vehicle garage, gas/petrol station, tannery, workshop in which wood is impregnated or worked)
Information
Topic
Table 5.4-4: Elements of a conceptual site model
topographic map, map of land use
Illustrations
Fig. 5.4-19
Fig. 5.4-17
Example figures/tables
964 5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
integrated interpretation and assessment of the model elements
mineralogical, geochemical and microbiological conditions
aquifer(s)/aquiclude(s), aquifer/aquiclude properties, groundwater table, soil water water well records (tables), content, hydraulic heads, direction and rate of groundwater flow, groundwater recharge well-construction diagrams, groundwater table map(s) and discharge (area, type, rate), importance of the aquifer(s) (contours/flow directions for various aquifers, if different), water quality table(s), water quality map(s) (water type, concentration isopleths) composition and properties of soil, rock and groundwater, estimation of contaminant map of contaminant sources, maps of contaminant retention, microbial activity, natural attenuation distribution and environmental parameters, tables of mineralogical, physical and geochemical analyses
major geological structures, stratigraphy and lithology, thickness and lateral extent of geologic column(s), well geological units, stratigraphy, lithology, homogeneity and heterogeneity, bedding log(s), geologic map showing dip, strike, folds, faults, conditions and tectonic structures, fractures, impact of weathering, soil structure map(s), crosssection(s) streams, lakes and ponds, springs, wells, use and quality of surface water and map of streams, lakes and other surface water, groundwater, runoff, water balance hydrograph(s),
geological setting
hydrologic and hydrogeological conditions
relief, major landforms topographic map, aerial photographs, ground-level photographs
map of natural hazards
geomorphology
geotechnical stability
human activities: mining damage, buildings, quarries, gravel pits, clay pits, etc.
geogenic hazards: active faults, karst, earthquakes, subsidence, landslides, flooding
climate: precipitation, temperature, evapotranspiration, direction and velocity of the long-term average climatic wind water balance graph
Tables 5.4-6, 5.4-7, 5.4-8, 5.4-9
Figs. 5.4-28, 5.4-29, 5.4-30
Figs. 5.4-20, 5.4-21, 5.4-22, 5.4-23, 5.4-24, 5.4-25, 5.4-26, 5.4-27
Fig. 5.4-18
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
965
966
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
Hydrostratigraphic units are the layers of rocks with similar hydrogeological properties (FH-DGG, 1999). Geological cross-sections are constructed on the basis of the results of geophysical measurements, remote sensing and drilling investigations. These provide an impression of the spatial distribution of geological units. These geological sections must be transformed into hydrostratigraphic sections using hydrogeological properties and stratigraphic information for correlation. Schemes for the classification of hydrostratigraphic units have been developed for the main geological units by a working group of the German State Geological Surveys. As an example, the classification of the unconsolidated Quaternary and Tertiary rocks of the North German Basin is given in Table 5.4-5. Layers are grouped into hydrostratigraphic units according to they are considered to be aquifers or aquicludes. The next step in the development of a conceptual model of a site is to map the distribution of aquifers and aquicludes. Knowledge about hydrogeological connections (windows) between aquifers is especially important, because in these places contaminants can spread to deeper aquifer(s). Such connections can be caused by changes in lithology or tectonic activity (see for example Figs. 6.2-25, 6.2-5, 6.2-9, and 6.2-10). Geophysical measurements can provide helpful information about hydrogeological windows (see Parts 4 and 6). Not only the spatial distribution of the hydrostratigraphic units, but also the representative properties of these units must be determined. Rock genesis, lithofacies distribution, tectonic evolution of the structures must be taken into account for quantification of the properties of hydrostratigraphic units. In some cases (for example in colluvial sediments), it may be impossible to correlate aquifers and aquicludes over a large area owing to their heterogeneous lithology. Heterogeneous aquifers can be modeled using stochastic methods which use statistical values and probability laws to generate possible parameter distributions. An initial evaluation of the groundwater dynamics is done by preparing groundwater equipotential contour maps derived from groundwater levels in observation wells over the period of a hydrological year (including both dry and wet seasons). It is important for the screen depths of the observation wells to correspond to the hydrostratigraphic units. Construction of groundwater isohypses using piezometric head data from different aquifers is a common error. Other errors result from not knowing or incorrectly interpreting the interaction between groundwater and surface water flow.
967
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
Table 5.4-5: Hydrostratigraphic units of the unconsolidated Quaternary and Tertiary rocks of the North German Basin, after MANHENKE et al. (2001); F – aquifers, C – aquicludes Hydrostratigraphic Lithology Lithogenetic/lithostratigraphic unit (predominant) units C1 peat, silt, loam raised bog and fen peat, turfy moulder, meadow loam, sapropel F1 sand, gravel flood-plain sand, dune sand, blown sand, river gravel, lowland sand, melt-water deposits, calcareous tufa C2 till, loess Vistulian ground moraines, periglacial deposits F2 sand, gravel melt-water deposits of the Saalian detrital subsequent stage to foreset phase of the Vistulian, including calcareous tufa C3 tell, silt, clay Saalian ground moraines, glaciallake deposits F3 sand, gravel melt-water deposits of the Elsterian detrital subsequent stage to Saalian foreset phase, river gravel in middle-terrace deposits, calcareous tufa C4 silt, clay, till Lauenburg Clay Elsterian ground moraines, glacial-lake deposits F4.1 sand, gravel melt-water deposits, river gravel, middle-terrace deposits, younger and older main terrace deposits, upper-terrace deposits decomposed coarse gravel F4.2 sand, gravel melt-water deposits in deep channels F4.3 sand kaolin sands, micaceous fine sands of the Pliocene and Upper Miocene, sands of the Rauno strata C5 clay, silt, Upper Micaceous Clay, 1st and lignite 2nd Lusatian lignite horizons, silts of the Rauno strata F5 sand Upper Lignite Sands/marine sands, sands of the Oxlund strata, Lower Brieske Sands
Stratigraphy Holocene Holocene, Pleistocene (Vistulian) Pleistocene (Vistulian) Pleistocene (Saalian to Vistulian) Pleistocene (Saalian) Pleistocene (Elsterian to Saalian)
Pleistocene (Elsterian) Pleistocene (Lower Pleistocene to Elsterian) Pleistocene (Elsterian) Tertiary (Pliocene, Miocene) Tertiary (Miocene) Tertiary (Miocene)
968
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
Under semiarid climatic conditions, in which distinct wet and dry periods occur, the hydraulic interaction of surface water bodies and groundwater changes during the year. The water level in a stream rises during periods of intense precipitation. In this case the stream loses water, which means that surface water infiltrates into the aquifer. These conditions are called influent (losing stream) conditions. In periods with normal or low stream water levels, groundwater will discharge into the river, forming the base runoff of the stream. These conditions are called effluent (gaining stream). Figure 5.4-13 shows the influent and effluent conditions of a stream in an area with a high groundwater table. But over the course of a year, not only does the hydraulic interaction of surface water and groundwater change, but the groundwater table also varies as a result of changing groundwater recharge conditions. The changes in the groundwater levels are registered in hydrographs of observation wells. The groundwater table rises during recharge periods and declines at other times. The rise of the groundwater table can result in the leaching of waste or of other contaminated materials from landfills and other contaminated areas. Figure 5.4-14 shows a groundwater hydrograph of an observation well near a landfill in Thailand. During the dry period the groundwater table is below the bottom of the landfill, in the wet season the groundwater table rises to a level higher than the base of landfill. The result is leaching of substances from the waste. Withdrawal of groundwater from water wells, different recharge conditions in different years, human activity (irrigation, drainage) and other artificial and natural factors must also be taken into account when the groundwater dynamics is analyzed. Influenced by these factors, not only can the direction of groundwater flow change, but also the pressure gradients and, therefore, the recharge – discharge conditions between the aquifers.
Fig. 5.4-13: Influent (losing stream, blue) and effluent (gaining stream, orange) conditions of a stream in an area with a high groundwater table
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
969
Fig. 5.4-14: Groundwater hydrograph of an observation well near the Mae Hia landfill of Chiang Mai, Thailand
The boundary conditions are derived from hydrogeological data. Boundaries can be faults, facies changes or changes in hydraulic conditions (for example, flow lines, groundwater divides). Determination of the hydraulic conditions at the margins of the model and features within the model area which affect the hydraulic regime is required for the modeling of groundwater flow and contaminant transport. The reliability and the accuracy of the model largely depend on the definition of the boundary conditions. Three types of boundaries are commonly used (see also Section 5.4.3 and, for example, CASTANY & MARGAT, 1977): x a specified head or constant head boundary, x a specified flux boundary, and x head dependent boundary. A boundary with a measured piezometric head is called a specified head boundary. Lakes, ponds, streams, and ditches are typical specific head boundaries of a conceptual model. For groundwater modeling, specified head boundaries are more desirable than specified flux boundaries because head data can be measured much easier than flux data (USACE, 1999). A specified head boundary in a flow model represents an unlimited supply of water at this location. This also means that it has to be checked whether the flux passing
970
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
through the boundary in the flow model fits the flux of water available in reality. Specified flux boundaries are boundary conditions which are characterized by known, measured or estimated flux across the boundary. Pumping and injection wells, springs, surface infiltration, and leakage from a confining layer are typical elements. A special case of the specified flux boundary is a “no-flow” boundary, where no flux occurs. A “no-flow” boundary can be an impermeable barrier or model hydraulic boundaries such as groundwater divides or flow lines (“parallel flow” boundaries). When groundwater divides and flow lines are used as the outer boundaries of a model, care must be taken because these lines can move with the seasons or when the aquifer is under stress, e.g., during pumping. A head dependent boundary is present when the flux through the boundary is determined by a hydraulic conductivity term. An example of a head dependent boundary is a losing stream. The hydraulic head in the aquifer below the stream is dependent on the hydraulic conductivity of the stream bed, its thickness, and the head difference between the stream and the aquifer (ANDERSON & WOESSNER, 1992). The rate of groundwater recharge, together with the inflow and the outflow of the aquifer, is a governing factor in the groundwater balance. In the same way as the aquifer properties, the boundary conditions and the elements of the groundwater balance must be quantified in space coordinates. Before a numerical model can be prepared (Section 5.4.3), the refined conceptual hydrogeological model has to be transformed into a final conceptual or “information adequate” model. This involves a generalization of the model and assigning parameter values to the individual hydrogeological model units. How detailed an “information adequate” model needs to be depends on the objectives of investigation and on the quality of the data. Figure 5.4-15 schematically shows two possibilities for generalizing a hydrogeological model. In the one case (case A), a rough generalization is made, which can be sufficient to estimate, for example, the potential withdrawal of wells from two main aquifers. This is not sufficient, however, to assess the risk emanating from a contaminated area or landfill. In this case, the units and parameters of the hydrogeological model must be defined as precisely as possible, as done in case B. The procedures for discretization, calibration, and verification of a numerical model are described in Section 5.4.3.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
971
Fig. 5.4-15: Possibilities for the transformation of a hydrogeological model into an “information adequate” model, example from PETERS (1972)
For sites suspected to be hazardous, chemical data must be included in the conceptual hydrogeological model. A hazard ranking system (e.g., Hazard Ranking System (HRS) for uncontrolled waste sites from the U.S. EPA, http://www.epa.gov/superfund/programs/npl_hrs/hrsint.htm) uses information from screening level investigations to assess the relative potential of a site to pose a threat to human health and/or the environment. The HRS uses a structured analysis approach to assess a site. The computer program Quickscore http://www.epa.gov/superfund/programs/npl_hrs/quickscore.htm can be used to develop a conceptual model (collection of exposure pathways)
972
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
for site assessment, to calculate “scores”, and to identify gaps in the data. Scores are calculated for each pathway separately and are then combined using a root-mean-square equation to determine the overall score for the site. This ensures that the site score can be relatively high even if only one pathway score is high. Conservative assumptions are used for screening-level ecological risk assessment. Example: Conceptional model for the abandoned Nong Harn municipal waste disposal site near Chiang Mai, Thailand 2 The abandoned Nong Harn municipal waste disposal site is about 18 km north of the city center of Chiang Mai in northern Thailand. The site served the Chiang Mai Municipality and nearby sanitary districts as a central waste disposal facility from 1995 - 1998. The site is a former borrow pit for road construction material (clay, sand and gravel) with a lateral area of about 150 u 150 m and rather steep walls down to a depth of 40 - 50 m. The standards of that time were used for the planning and construction of the landfill, including HDPE liners, leachate collection and treatment systems as well as compactors (see Fig. 3.1-1). Consequently, the site was considered suitable for the disposal of a wide variety of waste types, including industrial waste. To the southwest of the Nong Harn landfill there is another smaller, older landfill, also in a former borrow pit.
Fig. 5.4-16: Nong Harn landfill in 2002 2 Condensed version of the report by GRISSEMANN et al. (2006).
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
973
But, the information about how completely the HDPE liner covers the Nong Harn landfill is contradictory. The drainage system for the seepage water does not function properly. The originally constructed leachate collection system has been broken by the settling of the waste. According to the available information, mainly municipal household waste has been disposed of. The typical composition of municipal solid waste in Chiang Mai in 1995 is shown in Fig. 5.4-17. It should be noted that the main component is organic waste, which can amount to 70 % of the total waste. The general moisture content was as high as 54 %. At that time, 250 tons of waste per day were generated in Chiang Mai, but only 200 tons per day were collected. A low degree of compaction is assumed. To avoid odor emission, the waste was covered with layers of clay. According to the Pollution Control Department (PCD) in Bangkok, this covering was not done on a daily basis nor was it done over the entire surface. The amount of cover material is not known but roughly estimated to amount up to one-third of the volume of the landfill. The volume of the landfill was calculated to be 530 375 m3. Due to settlement of the waste, the central part is lower than the margins of the landfill and a small pond several meters deep formed, covering approximately one-fifth of the landfill (Fig. 5.4-16). The pond overflows in the SW corner of the landfill during rainy season. Intense degassing occurs from the surface of the pond, along fracture zones at the margins of the landfill, and from degassing wells on the site. There are twenty degassing wells in the landfill. The main objective of this investigation was a risk assessment based on a comprehensive geological and hydrogeological model. The following work was carried out for the investigation: x compilation of information from available geoscientific maps, reports, satellite images, air photos, field observations, archive material and interviews, preparation of a base map, x implementation of geophysical field surveys for an assessment of the geological structure and the delineation of more detailed follow-up investigations, x drilling and completion of observation wells; well logging and sampling of cores for laboratory investigations, hydraulic well tests, x monitoring of the wells and groundwater sampling for chemical analysis, x integrated data interpretation, assessment of the risk potential, recommendations for follow-up investigations and pollution prevention measures.
974
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
Fig. 5.4-17: Typical waste composition for Chiang Mai, TATONG (1997)
Climate: Annual rainfall in the area ranges from 800 mm in the lowland to more than 1500 mm in the surrounding mountains. The main period of precipitation is the rainy season between May and October with maximum values of more than 200 mm in August. Temperatures vary between a minimum of about 13 °C in January and a maximum of about 36 °C in April. The annual mean temperature lies around 25 °C (1951 - 2000). Potential evaporation exceeds rainfall, except between July and September (Fig. 5.4-18). This is the main period of groundwater recharge (MARGANE et al., 1999a).
Fig. 5.4-18: Long-term average (1961 - 90) climatic water balance of the area around the city of Chiang Mai according to data provided by the Chiang Mai Airport weather station
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
975
Geological setting: The Nong Harn landfill is at the eastern rim of the Chiang Mai-Lamphun Basin, which is interpreted as a fault-controlled west-dipping half-graben system. Tectonic subsidence of the intramontane Chiang MaiLamphun Basin is believed to have started in the late Cretaceous/early Tertiary (TANTIWANIT & DORN, 1999; DORN & TANTIWANIT 2002). The surrounding mountains consist of Precambrian metamorphic and plutonic rocks as well as strongly consolidated Paleozoic sediments with interbedded volcanics. The basin fill, up to 2000 m in its deepest parts (WATTANANIKORN et al. 1995), consists of predominantly clastic rocks of fluviatile, alluvial and colluvial origin of Quaternary to probably also upper Tertiary age. Colluvial sediments occur on the mountain slopes and make up the foothills at the basin margins. These slope-derived sediments of Quaternary age are well exposed in numerous borrow pits as clay rich, poorly sorted material of mainly reddish color with intercalated silty-clayey sand and gravel beds. Gray-colored and more sandy horizons with erosional washout structures are also visible. Towards the basin center, the slope-wash deposits grade into alluvial and fluviatile sediments of the Ping River system. Though the colluvial sediments are young in age they are remarkably well consolidated. This widespread phenomena is best documented in borrow pits where the material has been excavated partly down to 50 m depth along nearly vertical walls. It is assumed that the colluvium is underlain by deeply weathered basement rocks of Paleozoic age which show a fault controlled dip towards the basin center. Volcanic rocks are shown to be present at depth in the Nong Harn area by the presence of outcropping basalts in the southeastern vicinity of the site. Various types of alluvial deposits occur between the landfill and the Ping River to the west. Besides flood plain deposits and unspecified alluvial deposits, upper terrace gravel beds of the Mae Taeng Group are part of these sediments. The terraces consist mainly of sandstone, quartz and quartzite gravel and pebbles, which are generally well rounded. Coarse gravel horizons are frequently interbedded with medium to fine-grained sands, silts or silty clays. General groundwater situation: The local drainage pattern in the area of Nong Harn is directed towards the Ping River in the west, which acts as the central drainage channel of the Chiang Mai-Lamphun Basin. The water table in the area around the landfill is at a depth of several tens of meters. For this reason, there are no dug wells, which are widely used in other parts of the basin. Water supply from groundwater is limited to deep wells below 50 m, and numerous rain-filled pits serve as reservoirs.
976
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
Fig. 5.4-19: Aerial photograph of the area around the Nong Harn landfill, together with information on geographic features
Results of the site investigation An interpretation of aerial photographs is shown in Fig. 5.4-20. The map shows a set of several lineaments in the area around the landfill site and three main directions of linear features and geophysical indications of faults crossing the survey area. At the site a combined ground geophysical survey was carried out with reflection/refraction seismic, gravity, magnetic, electromagnetic and dc resistivity methods (Geophysik GGD 2002). The geological structures in the area around the Nong Harn landfill are best documented in the seismic survey. For reasons of clarity, the seismic section in Fig. 5.4-28 does not show interpreted tectonic elements but only focus on the seismic sequences that have characteristic reflection patterns. Four seismic sequences can be distinguished within the investigation area. Sequences A and B show a slightly prograding reflection pattern to SSW, separated by an unconformity (red line). The lithologies of sequences A, B and D are documented in three wells close to the landfill (MW-NH1, MW-NH2, MW-NH3), which were completed as observation wells.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
977
Fig. 5.4-20: Structure map as derived from interpretation of aerial photographs and geophysical data
978
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
Well logging and cuttings indicate sequence A to be predominantly clay-rich with interbedded sand horizons while sequence B is more sand-rich with clayey intercalations and less consolidated. Sequence D is mainly made up of red clays which become progressively harder (low drilling rate) down to 98 m. They can be considered as “basement rocks” with respect to the overlying and less consolidated colluvial and alluvial sediments. Sequence C does not seem to exist directly below the landfill. Line NH0201 (Fig. 5.4-28) indicates that this sequence wedges out against sequence D, which rises steeply towards the east and might reach the surface at some distance east of the site. Interpreted fault patterns along this line suggest local uplift of sequence D beneath the landfill. Genetically, sequence C appears to consist of slope-derived colluvial sediments from the nearby mountains. In addition to the seismic survey, the gravity and magnetic surveys provide further information about the geology of the area around the Nong Harn landfill, in particular the structure of the basement. Combining the results of all three methods, it was possible to produce a structure map of the investigation area (Fig. 5.4-21). According to this map, the geology can be expected to be fairly complicated. A number of local gravity highs (redbrown) and lows (green) are observed. The gravity gradients which mark faults and structure boundaries or material changes trend in very different directions and show a number of intersections. The main regional trend is NW-SE but other directions (e.g., E-W or NE–SW) can be found too. This means that there has been considerable tectonic activity and erosion in the area. The highest elevation in the area is northwest of the landfill. The landfill is at the north edge of a local gravity high, which possibly marks a block in the ground. The almost E-W-trending southern gradient of this gravity high can be related to a fault indicated in the seismic data. A small graben structure found on seismic profile NH0201 southwest of the landfill cannot be resolved by the gravity measurements, probably due to the large spacing between stations. The local gravity high directly south of the landfill also seems to be relevant with regard to the groundwater flow. Considering the surface topography, groundwater movement towards the south would be more likely than towards the north. But, the groundwater investigations show that the groundwater flows to the northwest, which correlates with the observed gravity gradient. Assuming a bulk density of 2.1 g cm-3 for the unconsolidated sediments above the siltstone found in boreholes NH01 and NH02 and a bulk density of 2.4 g cm-3 for the siltstone, the gravity gradient northwest of the landfill can be modeled as the dipping surface of the siltstone. The depth of the siltstone calculated by gravity modeling in the eastern part of the seismic profile NH0101 correlates well with the depth of the basement determined by the seismic survey (about 160 m below ground surface).
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
979
Fig. 5.4-21: Structure map of the Nong Harn area based on geophysical data: residual gravity field from high-pass wavelength filtering with a cut-off wavelength of 2 km
980
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
The central gravity high continues to the northeast of the landfill, but its amplitude is lower than it is south of the landfill. Comparing this gravity anomaly with seismic profile NH0201 (Fig. 5.4-28), it can be seen that a facies change correlates with a gravity gradient. Northeast of this gradient, the seismic results indicates the top of the basement is at about the same depth as southwest of it but that it consists of a different material (facies II). According to the gravity anomaly, this material should have a lower bulk density than the basement material in the area of the landfill. The extension of facies II towards the northwest can also be seen in the gravity field and the fault indicated by the seismic results. The northwestern boundary of this block is confirmed by a gravity gradient trending NE-SW. Another larger fault structure, also marked on the geological map, is expected to be related to the NNW-SSE-trending gravity low at the northeastern margin of the survey area. This zone corresponds to a small valley in this area. Additional information about the basement was expected to be obtained from the magnetic survey, especially as the occurrence of basaltic rocks was indicated on the regional geological map. Except in the central part of the investigation area, the magnetic field does not significantly change – only about 20 nT over a distance of 3 km. The values increase from the valley to the mountains. The regional trend of this weak magnetic gradient corresponds to that of the regional gravity gradient. On the basis of this agreement, it may be assumed that both regional trends are related to the basement topography. Due to the fact that just part of an extensive magnetic anomaly was surveyed and that the gradient of this anomaly is low, it is assumed that the magnetic anomaly is caused by magnetic material at greater depth (probably basaltic rocks). The other possible cause, material with low magnetization near the surface, is less probable as the residual gravity field shows local changes and not a smooth picture like the magnetic data. The geophysical data also provide essential information about the landfill itself. In particular, information about the influence of the landfill on the surrounding area was obtained from the 2-D resistivity survey. Gravity and magnetic modeling indicate the shape of the landfill and composition of the material in it. Electromagnetic measurements proved their usefulness to quickly determine the lateral boundaries of a landfill. The profiles of the 2-D resistivity survey were placed in such a way that they could provide information about the landfills and about possible contamination plumes – some profiles cross the landfills, some run close to both sites and some are at some distance to see if any influence of the landfill can be detected. Profile P2 (Fig. 5.4-22) crosses the center of the Nong Harn landfill from west to east. The location of the waste pit is clearly marked in the resulting section by very low resistivities. Different parts of the landfill show different resistivities. Slightly higher resistivities near the surface are related to the cover material and there are two regions with very low resistivities in the
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
981
interior. The boundaries of the pit are steeply dipping with no indication of significant lateral flux of water. Only at the eastern wall can leakage from the landfill be expected as the resistivity gradient in profile 2 is not inclined inwards as would be expected from the construction of the pit. Corresponding to the gravity data, the western part of the pit is either not as deep as the eastern or higher resistivity material is in this part of the pit (see Fig. 3.1-1). The depth of the pit (40 to 50 m) could not be determined with these measurements as the depth of penetration of the method (about 40 m) was too small. Profile 1 (Fig.5.4-23) extends west of the main landfill and east of an old landfill (see Fig. 5.4-19 for orientation). Low resistivities can be observed at the surface of the main landfill. This could be caused by conductive material at the surface either due to either waste disposal in shallow pits or depressions or water overflowing from the depression in the middle of the landfill or seeping from the sides during the rainy season. Down to a depth of about 20 - 25 m there is no indication of leakage from the pits, but below this depth, the decrease in electrical resistivity seems to be related to the landfills. It is assumed that seepage water is leaving the pits in this area. This assumption is supported by measurements in borehole NH3a, where groundwater conductivities of more than 1900 µS cm-1 were determined at the northern end of the profile. However, it can be seen that the influence of the landfill decreases rapidly, i.e. no strong contamination plume can be found in the direction of groundwater flow at the depth range investigated.
Fig. 5.4-22: Profile 2, 2-D resistivity survey across the Nong Harn landfill
982
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
Fig. 5.4-23: Profile 1, 2-D resistivity survey at the western margin of the Nong Harn landfill
As the depth of penetration of the 2-D resistivity survey was limited for technical reasons to about 40 m, resistivity soundings (VES) were also carried out to obtain information about the local groundwater system and to fill the gap between results of the near-surface 2D-resistivity survey and those of the seismic survey. Except for the VES in the vicinity of the landfill, almost all of the sounding curves are similar – lower resistivities at the beginning of the curve, high resistivities in the middle part and decreasing values for the large spacings, i.e. at greater depths. The low resistivities of the layers near the surface could be caused by weathered lateritic material. Below that there is a zone which is also above the groundwater level, but apparently not weathered, thus, causing high resistivities. Although a high content of cohesive material was found in the area around the landfill, the very high resistivities partly determined for this layer (up to more than 2000 : m), indicate that, at least locally, the content of sand and gravel is relatively high. Decreasing values at larger spacings are related to a higher content of cohesive material and/or to increasing groundwater saturation of the unconsolidated material observed in the drill holes. The main reason for the comparatively high resistivity values, even for clayey material, is the lack of water. In the area around the landfill, the depth of the groundwater was determined to be about 40 m, which was confirmed by the boreholes. As the depth of the Nong Harn waste pit was roughly known, it was possible to estimate the average bulk density of the waste from the gravity measurements. For this reason, detailed gravity measurements were made on profiles across the Nong Harn landfill and the older landfill southwest of it in order to obtain better data for modeling. A residual field was used for comparison with the model gravity (see upper left picture in Fig. 5.4-24). This field was derived from the BOUGUER gravity by removing a linear trend. In this residual field, the gravity lows caused by the waste material, which has a lower bulk density than the surrounding rock, can clearly be seen.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
983
The only input information for profile 1 across the Nong Harn landfill is that the pit was expected to be 40 - 50 m deep. Assuming a depth of 45 m, the calculated average density of the waste is about 1.70 g cm-3. This is a value significantly higher than for normal domestic waste (a1.5 g cm-3) or organic material (about 1.3 g cm-3). Whether this higher value is caused by the clay layers which were alternately deposited or denser waste material was also disposed of cannot be determined on the basis of the gravity data. From the shape of the gravity anomaly, it must be assumed that the eastern wall of the pit is much steeper than the western one. This result corresponds to the information that there was an access ramp on the western side. The same density for the waste in the old landfill and a depth of about 20 m for that pit was calculated from the data from profile 2 (Fig. 5.4-24). This density is in quite good agreement with the result of the 2-D resistivity measurements along two profiles, where the depth of the older pit was estimated to be 20 - 25 m. As strong magnetic anomalies were found in the area of the waste disposal sites, a more closely spaced grid was measured to define these anomalies more precisely and to possibly find an explanation for the elevated values, as it was not expected that a significant amount of iron was present in the waste. In Nong Harn there is a single clear anomaly with a minimum in the center and two maxima north and south. A similar anomaly is present in the area of the old landfill. The shape of the anomalies with its large E-W trend is typical of the Earth's magnetic field at the low latitude of northern Thailand. According to model calculations, such anomalies can be caused by the presence of iron material in the waste. The measured field is shown in the upper left of Fig. 5.4-25. A model magnetic body which would fit the measured anomaly on a N-S profile is shown on the upper right. The calculated anomaly caused by such a body is shown on the lower left. It can be seen that the shape of this model anomaly corresponds to the shape of the measured anomaly. As this rough estimate demonstrates, a body at a depth of about 30 m and a thickness of about 15 m with a susceptibility of 0.075 (cgs) would fit the measured anomaly. The amplitude of the anomaly is more than 2000 nT. An iron content of 0.1 to 1 % in the model body would be sufficient to produce such an anomaly.
984
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
Fig. 5.4-24: Gravity method – modeling of the waste pits
Fig. 5.4-25: Magnetic method – modeling of the anomaly related to the Nong Harn landfill
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
985
Three boreholes were drilled and geophysical borehole logging conducted to obtain information about the geology and the hydrogeology of the area. The material was recovered from one of the boreholes to obtain an accurate lithological description of the sediments, to correlate them with the geophysical well logs and for laboratory investigations. Three groundwater observation wells were installed in the area around the site to detect possible groundwater contamination and to determine the direction of groundwater flow. Eight shallow percussion core boreholes with a maximum depth of 4 m were drilled to obtain soil gas samples in order to detect near-surface gas migration. Groundwater samples were collected from the observation wells. The samples were filtered in the field, conserved in prepared bottles, saved in ice-cooled boxes and analyzed in laboratory. After well completion, pumping tests were carried out in each observation well to determine its hydrodynamic aquifer parameter values. Well MW-NH-1 was drilled to a depth of 98.60 m. The lithological profile and well logs are shown in Fig. 5.4-26. The lithological profile shows clayrich sediments in the top 33 m, with only little sandy intercalation. Sand and gravel predominates between 33 and 65 m. Clay is present from 65 to 88.5 m, followed by siltstone. Drilling progress slowed noticeably below 65 m and below 85 m penetration rates became increasingly slower. The lithological sequences found in the seismic survey are well represented by the results from this borehole. The sequences A (0 - 33 m b.g.l), B (33 - 65 m b.g.l.) and D (below 65 m b.g.l) in Fig. 5.4-28 were penetrated. In the interval from 13 to 19 m drilling mud was lost, indicating potential groundwater migration pathways in this depth range.
986
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
Fig. 5.4-26: Geophysical well logs in borehole MW-NH-1 and the lithological interpretation (right) based on the drill cuttings. Caliper measurements are limited to a diameter of 20 cm (AZTEC Engineering, Lampang, commissioned by WADIS-project).
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
987
The investigation area seems to be characterized by very frequent facial changes. Correlation of wells MW-NH-1 and MW-NH-2a shows that similar sediment sequences are found over short distances, despite the fact that colluvial sediments generally change properties within short distances. The results of lithological interpretation of boreholes MW-NH-1 and MW-NH-2a are shown in the cross-section in Fig. 5.4-27. In the top 30 m, the bedding is approximately horizontal and the layer thicknesses are similar. The lower depths are characterized by layers ascending to the east. Thickness of the sediments decreases in the same direction and the hard material (siltstone) is present at a depth of 60 m. This is in good accordance with the results from reflection seismic survey. In the area of the landfill, the seismic section NH0201 extends NE–SW. If boreholes MW-NH-1 and MW-NH-2a are projected onto this section, it can be seen that a fault crossing the profile causes uplift of the eastern part (or downslip of the western part) of the top of the basement. If this fault extends to the north, crossing the landfill, an alternative correlation pattern in the lower part of the profile (Fig. 5.4-27) has to be assumed. However, the attempt to connect fault indications on seismic profiles results in a tectonic pattern that shows this fault running NNW–SSE, not crossing the profile line connecting the two boreholes. Prior to drilling, two soil samples were taken from the unfilled pit just southeast of the landfill and three samples from the “big pit” approximately 2 km SW of the landfill and analyzed for sediment properties and mineral composition. Quartz is the predominant mineral in all of the samples. The proportion of material of grain size < 63 µm is especially high in the two samples from the unfilled pit (49 % and 65 %) and one sample from the “big pit” (59 %) (Table 5.4-6). These three samples show a very uniform grain-size distribution. This fact and the generally high proportion of silt and clay causes a low effective porosity and thus to practically impermeable sediment layers. Characteristic of all five samples is a medium effective cation exchange capacity (CECeff) of 1.0 - 2.8 cmol+/kg. They are little mineralized, but the eluate shows high nitrate and ammonium concentrations. All of the samples yield a strongly to moderately acidic solution when stirred with deionized water and none of the samples contain carbonate. All of the carbon is in organic carbon material.
988
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
Fig. 5.4-27: Schematic geological cross-section between the two boreholes MW-NH-1 and MW-NH-2a (vertical exaggeration)
989
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
Table 5.4-6: Soil parameters of samples from the area around the Nong Harn landfill; NH: samples from the unfilled pit to the southeast of the landfill; BP: samples from the “big pit” approximately 2 km southwest of the landfill Units Soil proportion < 2 mm proportion < 63 µm soil type
% %
pH electrical conductivity µS cm-1 total carbon (TC) % total organic carbon (TOC) % total inorganic carbon (TIC) % Effective CEC cmol+/kg pH (after adding BaCl2 solution) sum of exchangeable cations calcium cmol+/kg magnesium cmol+/kg sodium cmol+/kg potassium cmol+/kg iron cmol+/kg aluminium cmol+/kg Eluate after DIN 38414-4 (S4) ammonium mg L-1 nitrate mg L-1 o-phosphate mg L-1
NH1/01 NH2/01 BP1/01
BP2/01
BP3/01
44 14 silty gravel, fine to medium sand 4.93 5.13 54 21.2 0.021 0.025 0.020 0.029 < 0.005 < 0.005 1.3 1.0
100 59 silt, fine to medium sand
98 65 silt, fine to medium sand
100 97 49 27 silt, fine silty to sand medium sand
5.11 24.1 0.071 0.069 < 0.005 2.4
4.84 41.2 0.042 0.039 < 0.005 2.8
4.3
4.4
4.8
4.8
4.5
1.4 0.31 0.08 < 0.02 0.06 < 0.02 0.99
1.9 0.06 0.02 < 0.02 0.06 < 0.02 1.8
0.83 0.55 0.23 < 0.02 0.05 < 0.02 < 0.02
0.51 0.34 0.12 < 0.02 0.05 < 0.02 < 0.02
2.1 1.2 0.48 < 0.02 0.09 < 0.02 0.30
0.63 1.3 0.003
1.1 5.3 < 0.002
0.13 0.53 0.003
0.18 0.58 0.003
0.15 0.89 < 0.002
4.88 15.9 0.028 0.026 < 0.005 2.9
990
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
Table 5.4-7: Chemical parameters of sediments from liner core samples from the drilling of MW-NH-3a
Sediment Investigation proportion of > 2 mm fraction proportion of the < 2 mm fraction of the sample without cobbles pH electrical conductivity total carbon (TC) total organic carbon (TOC) total inorganic carbon (TIC) Eluate after DIN 38414-4 (S4) pH electrical conductivity dissolved organic carbon (DOC) SAK254 ammonium-N nitrate-N nitrite-N o-phosphate-P sulfate iron manganese zinc calcium magnesium sodium potassium * not measured
Units
NH3-3
NH3-6
NH3-10
NH3-11
%
31
20
23
30
%
42
62
56
46
µS cm-1 % % %
8.43 52.9 0.10 0.090 < 0.005
6.65 30.5 0.047 0.046 < 0.005
6.96 30.9 0.081 0.081 < 0.005
8.14 34.8 0.085 0.068 < 0.005
µS cm-1 mg L-1 m-1 mg L-1 mg L-1 mg L-1 mg L-1 mg L-1 mg L-1 mg L-1 mg L-1 mg L-1 mg L-1 mg L-1 mg L-1
7.83 13.5 1.5 n.m.* < 0.05 0.16 < 0.02 0.015 1.58 0.203 0.054 0.015 0.570 0.103 6.35 < 0.06
6.94 < 10 1.8 < 0.1 < 0.05 0.11 < 0.02 0.003 0.64 0.011 0.013 0.003 0.134 < 0.042 < 0.029 0.337
6.98 < 10 1.6 < 0.1 < 0.05 0.09 < 0.02 0.003 < 0.5 0.007 0.075 < 0.002 0.307 0.077 1.35 0.804
7.46 11.7 1.1 < 0.1 < 0.05 < 0.05 < 0.02 < 0.002 0.50 0.020 0.028 < 0.002 1.25 0.229 2.41 1.43
Further soil analyses were carried out on core samples recovered in a liner from borehole MW-NH-3a (Table 5.4-7). These samples were taken from a depth of 43 - 49 m. As geophysical well logging in borehole MW-NH-1 indicated contamination at a depth of 46 – 48 m, this interval was sampled in the nearby located drilling MW-NH-3a. Samples NH3-10 and NH3-11 were taken from this interval, while the upper two samples NH3-3 and NH3-6 come from an uninfluenced part of the aquifer. The analysis of these samples revealed no significant differences between the upper and the lower two samples (Table 5.4-7). The emission of landfill gas was determined on the landfill and the immediate surroundings. The gas was analyzed to determine its composition, the rate of seepage was determined, and the distribution of emission.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
991
Municipal waste contains a large amount of organic bound carbon in the form of food and garden wastes. Within a landfill, the organic carbon is transformed into gaseous substances by microorganisms. The resulting gas generally consists of up to 99 % methane (CH4) and carbon dioxide (CO2). The increasing gas pressure within the landfill results in diffuse migration of the gas leading to uncontrolled release of gas to the atmosphere if no measures are taken to collect the gas. Landfill gas poses a potential risk to health in different ways. The primary danger is the potential for the methane to form an explosive mixture with air. Mixtures containing 4.5 - 15 vol. % methane are explosive. Mixtures with a higher proportion of methane are flammable. The risk of explosion is present on the landfill itself as well as where gas is emitted in the area around the site. The gas emitted from the Nong Harn landfill (measured at points where it has not mixed with atmospheric air) consists of 57 - 61 % methane and 35 - 37 % carbon dioxide. This composition corresponds to steady-state anaerobic fermentation of waste with an exceptionally high organic content. The compounds H2S and NH3 were detected in landfill gas in trace concentrations of up to a maximum of 94 ppm (H2S) and 180 ppm (NH3). The wide range of organic substances observed in traces in the landfill gas was very much as expected. Remarkable is the composition of the gas in well W9, about 160 m NE of the landfill – 58 % methane and 28 % carbon dioxide, which is comparable to that of the gas seeping from the landfill. A very high rate of flow was measured at some of the main seepage points, which indicates a high gas production rate in the landfill. At well W9, the flow rate of 5 L gas per minute was measured. Hydrogeological investigations show that the geology of the area around the Nong Harn landfill correlates well with the occurrence of groundwater in the area. Water wells at and close to Wat Wiwek Waharam (a temple about 600 m west of the site), together with the evidence from the observation wells at the site and from water supply wells in the broad valley to the north, indicate that aquiferous horizons are present only in seismic sequence B. Wells which penetrated only sequence A remained dry. This situation is shown in Fig. 5.4-28, which displays an expanded part of seismic line NH0201 in the area around the landfill with projections of nearby wells. The 80 m deep borehole about 160 m NE of the site (see above) most likely reached sequence D, but did not find any groundwater. It only emitted methane gas, which obviously originated from the landfill. The aquifer presumably wedges out east of the landfill against the rather impermeable sequence D. Within the landfill the depth of groundwater table is believed to be near the bottom of the former borrow pit and therefore within its natural fluctuation interval. No wells are documented in the area, which penetrate sequence C.
992
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
Fig. 5.4-28: Expanded part of seismic line NH0201, length: 700 m
The water level in the unfilled pit southeast of the landfill is about 29 m below the surrounding ground level. The groundwater table in the area is at a depth of around 40 m, approximately the depth of the bottom of the landfill, varying several meters over the year (Fig. 5.4-29). Therefore, water in the unfilled pit, with a maximum depth of approximately 4 m, is well above the groundwater table and is derived solely from precipitation. Only wells MW-NH-2a, MW-NH-3a and MW-NH-E1 could be used to determine the direction of local groundwater flow. These wells have screens at depths between 45 - 49 m (MW-NH-2a and 3a), and a not exactly known section starting at approximately 30 m (MW-NH-E1). NW-NH-1 could not be used for this determination. The general groundwater flow direction between the wells at the edges of the landfill is to the NNW, as shown in Fig. 5.4-30. The difference in elevation of the groundwater table of 1.25 m over a distance of 106 m represents a groundwater gradient of I = 0.0118.
Fig. 5.4-29: Cross-section across the Nong Harn landfill site; wells and unfilled pit are projected onto the profile line. The depth of the bottom of the landfill is not exactly known and is shown here with an assumed depth of 45 m.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
993
Fig. 5.4-30: Local groundwater gradient between the observation wells at the Nong Harn waste disposal site on April 8, 2003. The groundwater gradient is indicated by the blue arrow.
Since the landfill is sealed with an HDPE liner, although incompletely, there is some limited hydraulic connection to the groundwater flow regime. Repeated water level measurements show that the direction of the gradient is rather consistent over all climatic seasons. The groundwater flow direction to the NNW seems to be in contrast to the surface topography of the area, which slopes slightly to the south. The gravity survey data showing a gravity gradient corresponding to the groundwater gradient might explain the direction of groundwater flow. The groundwater table in the area immediately surrounding the Nong Harn landfill is at a depth of approximately 40 m. The results of drilling show that the top 33 m consist of very clay-rich material with only thin layers of sand (Figs. 5.4-26 and 5.4-27). The main aquiferous horizons are below 33 m depth. Nevertheless, there are observations which indicate that perched water occurs in the upper layers during the rainy season. Pumping tests to determine hydraulic properties could be conducted only in well MW-NH-1. Wells MW-NH-2a and MW-NH-3a did not yield enough water to obtain a steady-state water level. Since well MW-NH-1 has screens at different depths, the determined kf value represents an average of all the screens. In a pumping test, the hydraulic conductivity of the aquifers was determined to be kf = 6.6·10-7 m s-1. However, while parts of the 20.8-m-thick aquifer have a low hydraulic conductivity, other sections will have a higher conductivity than the average value. The objective of the hydrochemical investigation was to determine whether landfill leachate from the Nong Harn landfill is present in the groundwater or not.
994
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
Since the leachate drainage system for the landfill is not functioning properly, it was not possible to collect a water sample from inside the landfill. Due to the high volume of gas production, drilling directly on the site was excluded. Samples were taken in concrete rings on the site that were part of the drainage system early in the operation of the landfill. Originally extending to the bottom of the landfill, but they have now collapsed due to settlement of the waste, the depths of these pipes are not exactly known. Leachate from inside the landfill reaches the surface of the water in these pipes as a result of convection driven by gas rising in the water and seen emerging at the water surface. Even though the water also contains rain water, thus changing concentrations of its contents, it can provide an indication of the original composition and of the contaminant contents. This water has a high electrical conductivity (1660 µS cm-1) relative to the background values and water from the observation wells. The water in the lake on the landfill and from well MW-NH-3a have similar conductivities. When these values are compared with those for leachate from other landfills, it is apparent that dilution is occurring. At the Mae Hia landfill, which was used for waste from Chiang Mai before Nong Harn, conductivities of up to 25 600 µS cm-1 were measured. Water taken from the concrete rings showed high chloride (282 mg L-1) and ammonium (50 mg L-1) contents, as well as organic compounds in low concentrations. Volatile organic compounds (VOCs) such as benzene, toluene, xylene and ethylbenzene were found in concentrations that make closer investigation recommendable, but should be at least monitored. Water samples were taken from the small lake on the landfill (NH-lake) and the unfilled pit (NH-open pit) southeast of the site. Besides the main ions, organic compounds were analyzed, especially with respect to toxicity and biodegradability. The water samples are alkaline (NH-lake) or slightly acidic (NH-open pit) and differed in their content of diluted and undiluted substances (Table 5.4-8). The NH-lake sample was more concentrated than the NH-open pit sample. Analysis for metals (As, Cd, Cr, Co, Cu, Ni, Pb, Zn) showed concentrations below the 0.1 mg L-1 (high limit because of the small samples). Fluorescence measurements showed that the humic contribution to the DOC is low. The NH-lake sample contained traces of resorcin and one other alkylphenol. Other investigations have shown that polycyclic aromatic hydrocarbons (PAH) are unlikely to be present. Furthermore, herbal protein, typical of phytoplankton bloom, was found in this sample. The NH-lake sample was rich in bacteria with a high diversity of species and a high abundance of plankton in, typical of eutrophic surface water. A luminescent bacteria test revealed no acute toxicity.
Environmental Geology, 5 Geological, Hydrogeological etc. Investigations
995
Table 5.4-8: Selected analytical results for surface water samples from Nong Harn Sample ID
Units
sample location sampling date temperature pH (sampling) electric conductivity chloride total organic content (TOC) dissolved organic content (DOC) biological oxygen demand (BOD) luminescence bacteria test
°C µS cm-1 mg L-1 mg L-1 mg L-1 mg L-1
NH-lake
NH-open pit
small lake on the landfill 17.04.01 34 8.5 1940 501 135 52 10.8 not toxic
unfilled pit south of the landfill 17.04.01 33 5 - 5.5 85.2 4.16 2.6 2.5 5.2 not toxic
Water samples were taken from the new observation wells MW-NH-1, MW-NH-3a, the existing observation well MW-NH-E1 and the production well PW-NH-2/40 at the nearby temple. The observation wells are near the landfill, the “temple” well is about 600 m west of the landfill. Before sampling, the observation wells had to be considered possibly affected by seepage water, whereas the temple well (PW-NH-2/40) represents natural background water. The samples showed pH values between 4.5 and 6 and a broad range of electric conductivities. While the water from the temple well and well MW-NH-1 have a low mineral content, higher concentrations were found in wells MW-NH-E1 and MW-NH-3a (Table 5.4-9). The sample from well MW-NH-3a shows the highest concentration of all main ions, with only chloride being exceptionally high (271 mg L-1). The iron and manganese concentrations are very high in the observation wells. Of the heavy metals, only cadmium in MW-NH-1 (not measured in all wells) was detected in amounts above the guideline values for drinking water (Thai Ministry of Industry). No bacteria were found in the analyzed samples. Phenolic substances were present in significant concentrations. While the concentrations in the well at the temple (PW-NH-2/40) are similar to background values in other parts of the Chiang Mai/Lamphun Basin and the Nong Harn area, concentrations in wells MW-NH-1 and MW-NH-E1 are clearly elevated, and especially so in well MW-NH-3a. Multiple sensor probe measurements were carried out continuously in well MW-NH-3 to determine groundwater parameter values. Several groundwater parameters, including water table fluctuations, were monitored in this way and found to correlate to some extent with other hydrogeological and hydrochemical data.
996
5.4 Interpretation of Geological, Hydrogeological, and Geochemical Results
Table 5.4-9: Selected parameters from analyses of groundwater samples from Nong Harn Well number Sampling date temperature pH EC O2 Eh total solids (TS) suspended solids (SS) biochemical oxygen demand chemical oxygen demand total organic carbon total inorganic carbon dissolved organic carbon phosphate phosphorus nitrite nitrogen (NO2-N) nitrate nitrogen (NO3-N) ammonia nitrogen (NH3-N) chloride (Cl–) sulfate (SO42-) magnesium (Mg2+) calcium (Ca2+) sodium (Na+) potassium (K+) boron (B) iron (Fe) manganese (Mn) zinc (Zn) aluminum (Al) arsenic (As) cadmium (Cd) chromium total (Cr) cobalt (Co) copper (Cu) fluoride (F-) molybdenum (Mo) nickel (Ni) lead (Pb) total coliform bacteria fecal coliform bacteria phenol MPN = most probable number
MW-NH-1 24 Apr. 2002 21 Oct. 2002 °C 27.3 - 29.1 4.5 - 4.9 µS cm-1 65 - 194 mg L-1 0.1 - 1.5 mV 205 - 265 mg L-1 128 - 216 mg L-1 2 - 12 mg L-1 < 0.60 mg L-1 < 4.0 - 4.0 mg L-1 0.46 - 0.88 mg L-1 18.9 - 54 mg L-1 0.56 mg L-1 0.01 mg L-1 0.05 mg L-1 0.14 - 0.28 mg L-1 0.02 - 0.6 mg L-1 45.5 - 47.9 mg L-1 8.6 - 15.1 mg L-1 0.86 - 4.6 mg L-1 3.7 - 14.9 mg L-1 5.6 - 35.4 mg L-1 0.5 - 2.84 µg L-1 < 100 µg L-1 100 - 1365 µg L-1 1 - 753 µg L-1 < 20 - 147 µg L-1 < 20 µg L-1 < 5.0 µg L-1 0.11 - 17 µg L-1 2 -2.6 µg L-1 0.2 - 6.2 µg L-1 1.2 - 2 µg L-1 < 0.01 µg L-1 0.34 µg L-1 6.2 µg L-1 1.4 MPN/100 mL